lOMoARcPSD|43275824

Econometrics Module 29052023

Introduction to economics (Haramaya University)

Scan to open on Studocu

Studocu is not sponsored or endorsed by any college or university

Downloaded by Biruk Bizuneh ([email protected])
 lOMoARcPSD|43275824

Introduction to Econometrics for

Management
First edition

[email protected]

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Table of Contents
Chapter 1: Introduction ....................................................................................................... 3
1.1 Definition, Scope, & Division of Econometrics .................................................. 3
1.2 Econometrics & Other Disciplines of Economics................................................ 3
1.3 Goals of Econometrics ......................................................................................... 4
1.3.1 Analysis: Testing of Economic Theories .................................................... 5
1.3.2 Policy Makings ........................................................................................... 5
1.3.3 Forecasting .................................................................................................. 5
1.4 Division of Econometrics ..................................................................................... 6
1.4.1 Theoretical Econometrics ........................................................................... 6
1.4.2 Applied Econometrics ................................................................................. 7
1.5 Methodology of Econometric Research ............................................................... 7
1.6 Exercise for chapter one ..................................................................................... 16
Chapter 2: Simple Linear Regression Model (The Classical linear regression model) .... 17
2.1. Introduction ........................................................................................................ 17
2.2. The simple linear regression analysis................................................................. 17
2.3. Assumptions of the linear stochastic regression model. .................................... 19
2.3.1. Assumption about Ui ...................................................................................... 19
2.3.2. Assumption about Ui & Xi ............................................................................. 20
2.3.3. Relationship about explanatory variables ....................................................... 20
2.4. Estimation of the model ..................................................................................... 21
2.5. Statistical tests of Estimates ............................................................................... 25
ˆ
2.6. The relationship between r2 and the slope coefficient  . .................................. 28
2.7. Significance test of the Parameter estimates ...................................................... 29
2.8. Exercise for chapter two ..................................................................................... 48
Chapter 3: Properties of the least square Estimates .......................................................... 49
3.1. The Gauss-Markov Theorem ............................................................................. 49
3.2. Importance of the BLUE Properties:- ................................................................ 50
3.3. Maximum Likelihood Estimation ...................................................................... 50
3.4. Exercise for chapter three ................................................................................... 54
3.5. Chapter 4: Multiple Linear Regression .............................................................. 55
4.1. Introduction ........................................................................................................ 55
4.2. Variance & standard errors of OLS estimators. ................................................. 57
4.3. Test of significance of the parameter estimates of multiple regressions............ 57
4.4. Importance of the statistical Test of Significance .............................................. 67
4.5. Exercise for chapter four .................................................................................... 68
Chapter 5: Relaxing the assumptions of the classical model ............................................ 69
5.1. Introduction ........................................................................................................ 69
5.2. Violation of the important assumptions ............................................................. 70
5.2.1. Hetroscedaticty:- ............................................................................................. 70
5.2.1.1. Consequences of hetroscedasticity ........................................................... 71
5.2.1.2. How to detect Hetroscedasticity ............................................................... 73
5.2.1.3. Solutions for Hetroscedasticity ................................................................. 77
5.2.1.4. How to remove Hetroscedasticity ............................................................. 93
5.2.2. Autocorrelation ............................................................................................... 95

Haramaya University 1 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

5.2.2.1. Graphic methods of detecting autocorrelation .......................................... 95

5.2.2.2. Consequence of autocorrelation .............................................................. 103
5.2.2.3. Sources of autocorrelation ...................................................................... 104
5.2.2.4. Solution for autocorrelation .................................................................... 105
5.2.3. Multicollinearity ........................................................................................... 117
5.2.3.1. Sources of multicollinearity .................................................................... 117
5.2.3.2. Consequence of multicollinearity ........................................................... 118
5.2.3.3. Solutions to the problem of multicollinearity ......................................... 123
5.3. Exercise for chapter .......................................................................................... 127
Chapter 6 : Estimation with Dummy Variables .............................................................. 129
6.1. Introduction ...................................................................................................... 129
6.2. Some important uses of Dummy variables ...................................................... 131
6.3. Use of Dummy variables in Seasonal analysis................................................. 135
6.4. Exercise for the chapter .................................................................................... 138
REFERENCE.................................................................................................................. 140

Haramaya University 2 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Chapter 1: Introduction
1.1 Definition, Scope, & Division of Econometrics
Econometrics deals with measurement of economic relationships between economic
variables (dependent & independent variables). The term econometrics is derived from
two Greek words. i.e. economy & measure. Different economists give different definition
for Econometrics, but all of them are arriving at the same conclusions and we can boils
down the whole definition in to the following. “Econometrics is the positive interaction
between data & ideas about the way the economy works.” The central role of
Econometrics was often regarded as one of estimating the parameters of the model as
efficient way as possible, given a particular set of data with which to apply statistical and
mathematical techniques. To test the validity of Economic theory Econometrics provides
us numerical values for the parameters of economic relationships and using these
numerical values we can verify the economic theories. To arrive at these numerical
values of economic relationships we use economic theory, mathematics, and statistics.
Though econometrics uses all these it is different from each one of them due to its
distinctive nature. One of the most distinctive natures of Econometrics is that it contains
the random term which is not reflected in mathematical economics & economic theory.

1.2 Econometrics & Other Disciplines of Economics

Econometrics is an integration of economic theory, economic statistics, mathematical
statistics and mathematical economics.

Economic theory: states a qualitative relationship between the explanatory and explained
variable using Cetrus Peribus assumptions.

Ex.1. Consumption depends up on current income (Yt) and previous income (Yt-1) of an
individual other things being constant. This theory does not give any insight how current
income and previous income will affect consumption by giving numerical values.

Mathematical Economics: It explains the economic theory in the equation or

mathematical forms. Or mathematical economics explain the theory of economics in to
mathematical relationship between variables. Let us explain the above theoretical
relationship in mathematical form /example 1/ as follows

Ex. 2:
Where
Ct: consumption expenditure
Yt: current income
Yt-1: previous income

Haramaya University 3 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Again this mathematical relation does not capture other factors that affect consumption
expenditure. Then mathematical economics explain the exact relationship between the
dependent variable (Ct) & the independent variables (Yt and Yt-1) by ignoring other
variables that affects consumption expenditure.

Economic Statistics:- It is a descriptive aspects of economic theory. i.e. by collecting,

processing and presenting economic data in the form of table & charts. Though Economic
statistics provides numerical data like mean, median - standard deviation etc. but it does
not make reliable the relationship between the economic variables.

Mathematical statistics:- This is based up on the probability theory, which are

developed on the basis of controlled experiments. This statistical method can be applied
in economic relationships because such experiment cannot be designed for economic
phenomena. This probability theory applied for very few cases in economics such as
agricultural or industrial experimentations.

In all of the above methods they completely ignore the other factors that will affect the
economic relationship but econometrics by developing a method for dealing with the
random term that will affect the economic relationships differentiate itself from the
remaining.

Ct =  + 1Yt +  2Yt − 1 + ut : − − − − − − − − − − (1.2)

All variables have the same meaning as equation (1.1)

Ut: means the random term which represents all other factors that will affect consumption
expenditure. These factors may be many such as, invention of new product, wealth, wind
fall gain & loss, migration, tradition, etc. are affecting consumption expenditure. All
these factors will have their own influences on the consumption expenditure. Then
econometrics by considering other factors (represented by Ui) will find numerical values
for coefficients of the variable that will explain the relationship to verifying economic
theories.

1.3 Goals of Econometrics

There are three main goals of Econometrics. These are:
➢ Testing of Economic theories /Analysis/
➢ Estimation of the coefficients of Economic relationship /Policy making/
➢ Forecasting the future values of economic magnitudes. /Forecasting/
Successful econometric application should include some combination of all the three
goals.

Haramaya University 4 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1.3.1 Analysis: Testing of Economic Theories

In the early stages of the development of economic theory, so called armchair economist,
were formulating basic economic principles using verbal explanation and applying
deductive procedure (from general to particular). During this period of time economists
by pure logical reasoning derive some general conclusions /laws/ concerning the working
process of economic system. In this period no attempt were made to examine whether the
theories explained adequately the actual economic behavior or not.

Now a day any economic theory is subject to the empirical test of econometrics. i.e. the
explanation power of economic theory to explain the economic behavior is tested by
econometrics. Then Econometrics aims primarily at the verification of Economic theories
& there by to know & decide how well they explain the observed behavior of the
economic units.

1.3.2 Policy Makings

This means by applying different methods of econometrics techniques we can obtain
individual numerical values for the coefficients of economic relationship. Using these
numerical values a decision can be undertaken by different economic agents.
Econometrics can supply MPC, elastic ties, MC, MR etc. Using these magnitudes
(numerical values) decision will be undertaken.
Ex.
D =  + 1 I +  2 Ex +  3 PI +  4 PEx + Ui ---------- - - - -1.3
Where
D= devaluation,
I - volume of import,
PE - volume of export;
PI - price of import,
PEx - Price of export.
Then devaluation will depend on all these explanatory variable coefficients. From these
coefficients we can have
B1 = Marginal propensity to import,
B2 = Marginal propensity to export
B3 & B4 import & export propensities respectively
Then on the basis of these coefficients of numerical values the government will decide
whether devaluation will eliminate the countries deficit or not.

1.3.3 Forecasting
It means using the numerical values of the coefficients of economic relationships we can
judge whether to take any policy measure in order to influence the future value of
economic variables or not

Assuming that the estimated result from the Ethiopian economy for the year 2005-2015.

Haramaya University 5 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Yˆ = −261.09 + 0.2453 Xi − − − − − − − − − − − − − − − − − − − −(1.4)

Where
Yi = Ethiopian expenditure on imported goods
Xti = Personal disposable income.
Then on the basis of the above result the government can able to know his expenditure in
any year after 2015 using the above equation. If disposable income (Xt) will be 1 million
in 2019 his expenditure on imported goods will be Ŷi = -261.09 + 0.2453 (1,000, 000) =
245,038.91 by the year 2019. Then since the government knows the future values of
expenditure on imported goods & services he can take any measure to increase or cut
down imports using these numerical values. Forecasting is used for both developed &
developing countries in different ways. i.e. developed countries used it for regulation of
their economics whereas developing countries used it for planning purpose.

1.4 Division of Econometrics

Just like any subjects Econometrics also decomposed in to two branches:- Theoretical &
Applied econometrics.

1.4.1 Theoretical Econometrics

It is the development of appropriate econometric methods for measuring economic
relationship between variables in theoretical Econometrics.
i. the data used for measurement purpose are observations from the real world but
are not derived from controlled experiments
ii. Econometric relationships are not exact.
The econometrics method that will be used in theoretical Econometrics may be classified
in to two
a) Single - equation techniques i.e. one side relationship between variables at a time.
Ex. Qd =  + 1 Pi + ui − − − − − − − − − − − − − (1.5)
Means quantity demanded depends up on the price of the commodity but not price
depends up on quantity. Then we have only one side causations. Then we can
apply Econometrics techniques only for this equation.

b) Simultaneous equation model :- When there is two sided causation

Ex. Equation (1.5) explains that quantity demand depends on the price of the
commodity but if the price of the commodity is in turn depends on the quantity of
commodity supplied. Then we will have two side causations
Pi =  + Qs. + Ui − − − (1.6)
Econometrics techniques will applied for three equations
Qd =  + pi + Ui − − − − − dd − equation − − − − − − − − − − − −(1.6)
Pi =  + qs i + ui − − − − − price − equation − − − − − − − − − − − (1.7)
Qd = Qs --------------------- Identity ---------------------------- - - - -(1.8)
Then in this case we applied Econometrics techniques simultaneously for all equations at
a time.
Haramaya University 6 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1.4.2 Applied Econometrics

This is the application of theoretical Econometrics methods to the specific branch of
economic theory i.e. application of theoretical Econometrics for verification &
forecasting of demand, cost, supply, production, investment, consumption & other related
field of economic theory.

1.5 Methodology of Econometric Research

In any econometrics research we may distinguish the following steps.

Economic theory

Mathematical model of the theory

Econometric model of the theory

Collecting data

Estimation of Econometric model

Evaluation of Estimates
(Hypothesis testing)

Application (forecasting)

Stage 1. The first step in econometrics is to formulate the economic theory that will be
tested against reality using econometrics.
Ex,
• The theory may hypothesize that "Aggregate saving in the economy is affected by
the average interest rate and the one year lag of income (Previous year income).
• Or if you take Keynes psychological law of consumption it hypothesize
consumption is a function of income to be precise. "Aggregate consumption in
terms of wage units (Cw) & aggregate income (Yw) in terms of wage units are
called this relationship propensity to consume.
• Or consumption of an individual at any period of time depends upon income of an
individual at period t and future rate of interest.
Stage 2. Specification of the Model:-This is transformation of econometric theory in to
mathematical model that explain the relationship between economic variables. Under this
stage we will have the followings.

Haramaya University 7 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

A) Selecting variables:- it involves the determination of dependent (endogenous or

explained) variable and independent (exogenous or explanatory) variables of the theory.
From the above example we can identify that
Ex.1: Aggregate saving is the dependent variable & the remaining variables (interest rate
& previous year income) are the independent variable.
E.x.2: Aggregate consumption in terms of wage units is the dependent variable &
aggregate income in terms of wage units is the independent variable
Ex. 3: Consumption of an individual at time t is dependent variable and future rate
interest & income are independent variables.

B) Determine the theoretical values: a prior expectation of the sign & magnitude of the
parameters. This needs only a theoretical background to determine the relationship
between the dependent & independent variables i.e. negative or positive relationship
between variables. From our example we can have the following sign or direction of
relationship between variables
Ex.1.

i. Interest rate & saving have positive relationship and also there is a positive
relationship between income and saving. Then we can say that the sign of the
parameters that will explain the relationship between aggregate saving & interest
rate & income have to be positive.

ii. Aggregate consumption in terms of wage & Aggregate income in terms of wage
are positively related & the sign of the parameter has to be positive

iii. The relationship between consumption at time t & income at time t has positive
relationship & consumption at time t & future rate of return have negative
relationship (if future rate of interest is high an individual will cut down his
consumption at time t & post pone his consumption for other period & increase
his savings).

C) Specification of the model: In this stage we specify the relationships between the
dependent & independent variables on the basis of economic theories. In this stage we
also determine the number of equations (single equation or simultaneous equation model)
& the type of equation i.e. whether the relationship between economic variables
explained using linear or non- linear equations. Let’s specify our previous theoretical
relationships.
Ex. 1.
St =  + 1Yt −1 +  2 r − − − − − − − − − − − − − − − (1.9)
Where; St = aggregate savings,
Yt-1 is previous income
r is rate of interest.
Ex- 2
Ct =  + 1Yw − − − − − − − − − − − − − − − − − (1.10)

Haramaya University 8 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Where; Ct = aggregate consumption in terms of wage

Yw is aggregate income in terms of wage.
Ex. 3.
Ct = Yi 1 ri  2 − − − − − − − − − − − − − − − − − −(1.11)
Where; Ct is consumption at time t,
Y is income at time t
re is future rate of interest.
All the above equation is single equation model but equation 1.9 & 1.10 are linear
equations & equation 1.11 is nonlinear equation. Magnitude of the coefficients of the
variables (  , 1 &  2 ), what will be the likely magnitudes of these coefficients? The
magnitude or the size of the numerical values of the coefficients of the variables
(  , 1 &  2 ) are determined by the economic theory & empirical observation of the real
world. In equation 1.9 & 1.10the coefficient 1 represents marginal propensity to
consume and the magnitude of 1 is 0<β<1 it is determined by the economic theory. The
explanation of the magnitude of equation 1.9 &1.10 are different from equation 1.11. In
equation 1.9 &1.10 which is a linear equation the coefficients of the variable explains the
marginal magnitudes but equation 1.11 explains the elastic ties.

Ex. In equation 1.9 & 1.10 if income increases by 1 birr on the average consumption will
increase by β amount. The same thing is true in the interpretation of β2 in equation 1.10
i.e. if rate of interest is increasing by one birr on the average saving will increased by β2
amount. But in equation 1.11 β1 & β2 explains elastic ties i.e. if income increases by 1%
consumption will increase on the average by β1% & for β2 if rate of interest is increasing
by 1% consumption will be cut down on the average by β2 %.

Possible errors committed at this stage are

➢ Wrongly specifying the model (i.e. relationships that will be explained using non-
linear relationship may be specifying in linear form & the reverse.) - Mis-
specification of the model.
➢ Relevant explanatory variables may not be included (Omission of relevant
explanatory variables)
➢ Irrelevant explanatory variables may be included.

Step - 3. Specification of the econometric model

The above equation (1.9 to 1.11) explains the mathematical relationship between the
explained & explanatory variables is inexact form or deterministic form. i.e. all the
dependent variable will be affected only by those independent variables alone & any
other variables cannot affect the dependent variable. But in reality different factors will
affect the economic relationships that will not be captured by our example.

Ex. In equation 1.9 states that saving will depend up on previous income & rate of
interest alone. But in reality many variables will affect savings such as wealth;

Haramaya University 9 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

consumption, windfall gain & loss, health of the individual, etc. There are many factors
that will affect the saving capacity of individuals. Then those factors which are not
incorporated in our model will make the relationship between the dependent & the
independent variables inexact. Then these factors make the Economic relationship
between the dependent & the independent variable is inexact & it can be specified in the
following model.

St =  + Yt −1 +  2 r + Ui − − − − − − − − − − - - - - - - 1.12
Cw =  + Yw + Ui − − − − − − − − − − − − − − − ------1.13
Ct = Yi 1 r − e 2 e ui − − − − − − − − − − − − − − − -------1.14
In all the above equations Ui represents all factors that affects the dependent variable but
not explained or taken in to account explicitly in the model. Then Ui is called the
disturbance term or error term or random term or stochastic variables. The inclusion of Ui
in the mathematical economics (exact relationship between variables) will transform the
model in to Econometric model (inexact relationship between variables or the
unexplained variable in the model will capture or explained by Ui.)

Step - 4 Obtaining Data

The data used in the estimation of econometric model may be of various types.

➢ Time series data: a data related to a sequence of observations over time on an

individual or group of individuals etc. Ex. 1996 E.C. represent by Ct where t indicates
time from 1980 - 1996 E.C.

➢ Cross-sectional data: data collected on one or more variables collected at particular

period of time. Ex. Number of children registered for schooling in all K.G. Schools of
Harar in 2014 E.C. by sex, age, religion etc.

➢ Panel data: - These are the results of repeated survey of a single (cross sectional data)
sample in different periods of time. Ex. If consumption expenditure of a sample of
population from Harar city on Teff, Coffee, cloth is taken in 2000, in 2007 & 2014.

➢ Polled data: - These are data of both time series & cross-sectional data.

➢ Dummy Variable: These are data constructed by econometricians when they are faced
with qualitative data. These qualitative data may not be measurable in any one of the
above methods ex. sex, religion, race, profession etc. The value of these data can be
approximated using dummy variables ex. if religion is appearing in the independent
side of the equation since we do not have qualitative data, we can assign 1 for
Christian & 0 Otherwise.

Haramaya University 10 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Accuracy of data: - Though plenty of data are available for research purpose but the
quality of data matters in arriving at a good results. The quality of data may not be good
for different reasons.
i. Most social science data are not - experimental in nature i.e. there will be
omission, errors etc.
ii. Approximation & round off the numbers will have errors of measurement.
iii. In questioner type of survey non-response and not giving an answer for all
questions may lead to selectivity bias. /rejecting non-response & excluding those
questions which is not answered by the respondent/
iv. The data obtained using one sample may be varying with the data obtained in
another sample & it is difficult to compare the results of these two samples.
v. Economic data are available at aggregated level & errors may be committed in
aggregation.
vi. Due to confidentiality of some data’s it is impossible to get the data or may be
published in aggregated form.

Because of the above reasons one can deduce that the results obtained by any researchers
are highly depending up on the quality of the data. Then if you get unsatisfactory results
the reason may be the quality of the data if you correctly specifying the model.

Step 5. Estimation of the Econometric Model

We can estimate the coefficients of the independent variables which explain the
relationship between economic variables in two different ways. Single or simultaneous
equation methods.
➢ Single equation method:- This techniques of estimation is applicable only for one
equation at a time
Ex. Qd=+ βPi+Ui-----------------------------1.15
Where, Qd= quantity demanded
P is price.
For this & any this kinds of equation we can apply different methods of estimation. These
are OLS (Ordinary least squares,) In direct least square or reduced form techniques, two
stages least squares (2SLS), Limited Information (LI), Maximum likelihood (MLI) &
Mixed estimation method may be used.

➢ Simultaneous equation techniques- When we have more than one equation & if the
numerical values of the coefficients are determined simultaneously at a time then we
use any one of the following methods of estimation, Three stage least squares (3SLS),
& the Full information Maximum Likelihood (FIML) method. The selection of the
techniques of estimation will depends upon many factors.

Haramaya University 11 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

a) Nature of the relationship between economic variables and its identification.

Under this condition if we studied the economic relationship using a single
equation method then the best method is OLS. But if the relationships between
economic variables are in a function of simultaneous equation we may use any
techniques from the above stated.
b) On the properties of the estimated coefficients obtained from each method a good
estimate should give the properties of unbiased ness, consistency, efficiency &
sufficiency or a combination of such properties. If one method gives an estimate
which passes more of these desirable characteristics than any other estimates from
other methods, then that techniques which possess more of the desirable
characteristics will be selected.
c) On the purpose of Econometric research: - If the purpose of the model is
forecasting the property of minimum variance is very important i.e. the techniques
which will give the minimum variance of the coefficients of the variables will be
selected. But if the purpose of the research is for policy making (analysis) that
techniques which gives unbiased ness of the variables will be selected.
d) On the simplicity of the techniques: If our interest is simple computation, we can
select that technique which involves simple computation & less data requirement.
e) Time & cost required for computation of the coefficients of the variables may
determine the selection of the Econometric techniques.
The estimation of the /coefficients of the variable/ the model can be computed using any
one of the above stated econometrics techniques. Some techniques which are
theoretically applicable may not be used for estimation purpose due to non-availability of
data or defaults of the statistical results obtained from the technique.

Having selected the econometric method that will be applicable for estimation of the
model one should take in to consideration whether the model is linear in variable & in
parameters.
a) If the model is non - linear in parameters it is beyond to this level of Econometric
analysis. Example:
Y =  + 12 X 1 + 13 X 2 + Ui -------------------------1.16
Since the coefficient β1 is the power of 2 & β2 is the power of 3 then we call these
kinds of model non-linear in the coefficients.
b) If the model is non-linear in variables then before estimation the model has to be
transformed in to linear model.
To know whether the model is linear or non-linear in variable we can take the first
derivatives & if the first derivative of the model gives us a constant number then
the model is linear in variables but if it doesn't give us a constant number the
model is non-linear in variable.
Example (1)

Haramaya University 12 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Yi =  + 1 X 1 + Ui − − − − − − − − − − − −(1.17)
yi
If you take the first derivation of Yi w.r.t. X. i.e = 1 [Which is the
xi
coefficient of Xi] is a constant number then the equation is linear in variable.
Example (2)
Yi =  +  X i 2 + Ui − − − − − − − − − − − (1.18)
yi
= 2X i Then we can say that the model is non-linear in variable because the
xi
first derivation w.r.t.x does not give us a constant number.
Example (3)
Yi = 1 +  2 ( 1 2 ) + Ui − − − − − − − − − − − (1.19)
X
yi −3
= −2 2 X i or −  2 ( 1 3 ) then since the first derivation is not equal to a
xi X
constant number then again, the model is non-linear in the variable.

To estimate the model which is non-linear in variable we should first transform the model
in to linear model.
Equation (1.18)
Yi =  + xi + Ui − − − − − − − − − − − − − − − −1.20
*

Where X i = x 2
*

Yi = 1 +  2 X i + Ui − − − − − − − − − − − − − − − 1.21
*
Equation (3)
Where X i = 1
*
xi
Again if you have the following models first transform as follows
−
Yt= xi e u transform in to lnyt= ln-  , ln Xi +Ui
Yt = e + xtu Transform in to lnyt =+βXi + Ui

e =  + x tui Transform in to y=+β log ex + Ui
−y

Y = e +  / xtui Transform in to y=+β ( 1 x ) + Ui

Having transformed the model from non-linearity in variable to linearity in variable then
we can estimate the model using the appropriate (selected) method of econometrics
methods.

Step - 6 Evaluations of Estimates:

After estimating & obtaining the coefficients of the variables one has to precede to the
evaluation of the results obtained using econometric methods. At this stage we are
evaluating the reliability of the results whether they are theoretically meaningful &
statistically satisfactory results.
To evaluate the reliability of the estimates we apply /follow/ the following steps
i. Economic interpretation of the results - Economic a priori criterion.

Haramaya University 13 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

ii. Statistical interpretation of the results - statistical analysis on the basis of R2 ,t,
test, F- test, s.d.
iii. Test of Econometric criterion.

Step A. Economic a prior criterion: at this stage we should confirm that whether the
estimated values explain the economic theory or not i.e. it refers to the sign & magnitudes
of the coefficients of the variables.
Ex. 1. If we have the following consumption function
Ct= +β1Yt + Ut -------------------------------------------1.22
Where Ct: consumption expenditure,
Yt is income
From the economic theory (economic relationship between consumption and income) it is
known that β represents MPC (Marginal Propensity to consume). Then on the basis of a
priori-economic criterion it is determined that the sign of β has to be positive & the
magnitude (size) β again is in between zero & one (0< β<1). If the estimated results of
the above consumption function gives
Cˆ = −3.32 + 0.2033Yt -----------------------------------1.23
i

From the economic relationship explained by economic theory states that if your income
increase by 1 birr your consumption will increase on average by less than one birr i.e .203
cents. Then the value of β1 is less than one & greater than zero in its magnitude (size)
again the sign of β1 is positive. Therefore, the estimated models explains the economic
theory (economic relationship between consumption & income) or satisfies the a priori -
economic criterion. If another estimation of the model using other data gives the
following estimated results
Cˆ = 24.45 − 5.091Yt -------------------------------------1.24
t
Where Ct is consumption expenditure Yt is income. From Economic theory it is known
that β1 has to be positive & its magnitude is greater than zero & less than one. But the
estimated model results that the sign of β1 is negative & its magnitude is greater than one
then we reject the model because the results are contradictory or do not confirm the
economic theory.

In the evaluation of the estimates of the model we should take in to consideration the sign
& magnitudes of the estimated coefficients. If the sign & magnitudes of the parameter do
not confirm the economic relationship between variables explained by the economic
theory then the model will be rejected. But if there is a good reason to accept the model
then the reason should be clearly stated. In general if the prior theoretical criterions are
not satisfied, the estimates should be considered as unsatisfactory. In most of the cases
the deficiencies of empirical data utilized for the estimation of the model are responsible
for the occurrence of wrong sign or size of the estimated parameters. The deficiency of
the empirical data means either the sample observation may not represents the population

Haramaya University 14 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

(due to sampling procedure problem or collecting inadequate data or some assumption of

the method employed are violated). In general if a priory criterion is not satisfied, the
estimates should be considered as unsatisfactory.

Step-B- First order test or statistical criterion: If the model passes a prior-economic
criterion the reliability of the estimates of the parameters will be evaluated using
statistical criterion. The most widely used statistical criterions are:
o The correlation coefficient - R2/r2/
o The standard error /deviation/ S.E of the estimate
o t- ratio or t-test of the estimates.
Since the estimated value is obtained from a sample of observations taken from the
population, the statistical test of the estimated values will help to find out how accurate
these estimates are (how they accurately explain the population?).
✓ R2 will explain that the percentage of the total variation of the dependent variable
explained by the change of the explanatory variables (how much % of the dependent
variable is explained by the explanatory variables).
✓ S.E. (Standard error or deviation) - measures the dispersion of the sample estimates
around the true population parameters. The lower the S.E. the higher the reliability
(the sample estimates are closer to the population parameters) of the estimates & vice
-versa.

Step -C- Second order test /Economic Criterion/: after testing a prior test & statistical
test the investigator should check the reliability of the estimates whether the econometric
assumptions are holds true or not. If any one of the assumption of econometric
assumptions are violated.
➢ The estimates of the parameters cease to possess some of the desirable properties (un
biasedness, consistency, sufficiency etc)
➢ Or the statistical criterion loses their validity & became unreliable.

If the assumptions of econometric techniques are violated then the researcher has to re –
specifying the already utilizing model. To do so the researcher introduce additional
variable in to the model or omit some variables from the model or transform the original
variables etc.

By re-specify the model the investigator proceeds with re- estimation & re-application of
all the tests (a priori, statistical and econometric) until the estimates satisfies all the tests.

Step 7. Forecasting or Prediction

Forecasting is one of the prime aims of econometric research the estimated model may
economically meaningful, statistically & econometrically correct for the sample period.
But given all these it may not have a good power of forecasting due to the inaccuracy of
the explanatory variables & deficiency of the data used in obtaining the estimated values.

Haramaya University 15 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

If this happens the estimated value (i.e. forecasted) should be compared with the actual
realized value magnitude of the relevant dependent variable. The difference between the
actual & forecasted value is tested statistically. If the difference is significant, we
concluded that the forecasting power of the model is poor. If it is statistically
insignificant the forecasting power of the model is good.

1.6 Exercise for chapter one

1. Define Econometrics? How does it differ from Mathematical Economics &
Mathematical statistics?
2. What are the goals of econometrics & explain it using example?
3. What is the difference between theoretical & applied econometrics?
4. What is the difference between the model which is linear in variable & & nonlinear in
variable? & how would you interoperate the parameters?
5. What are the steps applying to evaluate the reliability of the estimates?
6. Explain the difference between economic theory & econometrics?
7. Given the following theory which is given by the well-known monetary economist
Milton Fridman “the theory of demand for money have a strong positive relationship
with price & income but has no relationship with rate of interest”
➢ Write the mathematical relation ship
➢ Formulate the econometric relation ship
➢ What will be the size & magnitude of the relationship between the dependent &
independent variables?
8. Explain the stages in the methodology of econometrics?

Haramaya University 16 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Chapter 2: Simple Linear Regression Model (The Classical linear

regression model)

2.1. Introduction
In economics the relationship between variables are mainly explained in the form of
dependent & independent variables. The dependent variable is that variable which its
average value is computed using the already known values of the explanatory variable(s).
But the values of the explanatory variables are obtained from fixed or in repeated
sampling of the population.

Ex. Suppose the amount of commodity demanded by an individual is depend on the price
of the commodity, income of individual, price of other goods and so forth. Then from this
statement quantity demanded is the dependent variable which its value is determined by
the price of the commodity and income of the individual, Price of other goods etc. And
price of the commodity, income of individuals & price of other goods are independent
(explanatory) variables whose value is obtained from the population using repeated
sampling. The relationship between these dependent and independent variable is a
concern of regression analysis. i.e.
Qd = f (P, P0, Y etc) -------------------- (2.1)
If we study the relationship between dependent variable & one independent variable i.e.
Qd= f (P) this is known as simple two variable regression model because there are one
dependent Qd & one independent P regression model. However if the dependent variable
is depending upon more than one independent variables such as Qd: f (P, P0, Y) it is
known as multiple regression analysis. The functional relationship between the
dependent and independent variable may be linear or non-linear.

2.2. The simple linear regression analysis

The relationship between the dependent & independent variable suggested by economic
theory is usually specified as exact or deterministic relationships. But in reality the
relationship between economic variables are inexact or stochastic or in deterministic in
nature.
Ex. Suppose consumption expenditure for a commodity is depending up on current
income of the individual citrus-paribus & assumes that the functional relationship is
linear then we can write it.
Ct =  +Yt ------------------------------------------------------2.2
Then for each specific value of current income (Yt) there will be only one corresponding
value of consumption expenditure. This shows that consumption expenditure is depend
upon current income. But consumption expenditure is not only determined by income
alone but different variables such as wealth, previous income, tradition etc affect
consumption expenditure. Then there is inexact relationship between these two variables

Haramaya University 17 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

and to capture those factors which affects consumption expenditure in equation 2.2 we in
corporate a variable ‘U’. Then we can write the equation as follows
Ct =  +Yt + Ut----------------------------------------- (2.3)
Where Ct is the dependent variable,
Yt is independent variable,
 &  are regression parameters,
Ui is the stochastic disturbance term or error term.
We introduce ‘U’ – random term due to the following reasons.
i. Omission of variables from the function. In economic reality each variable is
influenced by very large number of factors and each variable may not be included
in the function because of
a) Some of the factors may not be known.
b) Even if we know them the factors may not be measured statistically example
psychological factors (test, preferences, expectations etc) are not measurable
c) Some factors are random appearing in an unpredictable way & time. Example
epidemic earth quacks e.t.c.
d) Some factors may be omitted due to their small influence on the dependent
variables
e) Even if all factors are known, the available data may not be adequate for the
measure of all factors influencing a relationship
ii. The erratic nature of human beings:- The human behavior may deviate from the
normal situation to a certain extent in unpredictable way.
iii. Misspecification of the mathematical model:- we may wrongly specified the
relationship between variables. We may form linear function to non- linearly
related relationships or we may use a single equation models for simultaneously
determined relationships.
iv. Error of aggregation: - Aggregation of data introduces error in relationship. In
many of Economics data are available in aggregate form ex. Consumption,
income etc is found in aggregate form which we are added magnitudes referring
to individuals where behavior is dissimilar.
v. Errors of measurement:- when we are collecting data we may commit errors of
measurement
In order to take in to account the above source of error we introduce in econometric
functions a random term variable which is usually denoted by the latter U & is called
error term, random disturbance term or stochastic term of the function. By introducing
this random term variable in the function the model will be just like equation number
(2.3). The relationship between variables will be split in to two parts.
Ex. From equation (2.3)
+Yt represents the exact relationships explained by the line

Haramaya University 18 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Apart represented by the random term Ui is the unexplained part by the line. This
can be explained using the following graph.
Y
Yn
Ct= =  +yt
Un
Ct Consumption

Yn

U2

Y1

X
Current income (Yt)
Figure 1: Regression line

The line Ct =  +Yt shows /explain/ the exact relationship between consumption &
income but other variables that affect consumption expenditure are scattered around the
straight line. Then the true relationship is explained by the scatter of observations
between Ct & Yt.
Ct =  +Yt + Ut ------------------------ (2. 4)
Variation in Explained
consumption+ + Unexplained
variation variation

To estimate this equation we need data on Ct, Yt &Ut, since Ut is never observed like
other variables (Ct & Yt) we should guess the value of ‘U’, that is we should make some
assumptions about the shape of each Ui (mean, S.E, Covariance etc)

2.3. Assumptions of the linear stochastic regression model.

To guess the value of ‘U’ we make some assumptions about Ui & divided these
assumptions in to three
a) Some refer to the distribution of random variable Ui.
b) Some to the relationship between Ui & the explanatory variables
c) Some refer to the relationship between the explanatory variables themselves

2.3.1. Assumption about Ui

i. Ui – is a random real variable:- The value which Ui-may assume in any one period
depends on chance, it may be positive, negative or zero.
ii. The mean value of Ui in any particular period is zero i.e.

Haramaya University 19 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

E(Ui) = 0 or  ui = 0
i =0
iii. Homoscedasticity: (Constant Variance). The variation of each Ui around all values of
the explanatory value is the same i.e. the deviation of Ui around the straight line (in
figure 1) is remain the same var (Ui)=  u
2

iv. The variable Ui has a normal distribution with mean zero & variance of Ui.
Ui is N(0,  u )
2

v. Ui is serially independent:- the value of U in one period is not depend up on the value
of Ui in other period of time means the co-variance between Ui & Uj is equal to zero
Cov (UiUj) = 0

Cov (UiUj) = E [ Ui – E (Uj)] [Uj –E(U)]

By assumption ii – the E(Ui) = 0 then

= E [Ui-0] [Uj-0]

= E(Ui) E(Uj)

Again by assumption E(Ui) = 0

Cov (UiUj ) = 0

2.3.2. Assumption about Ui & Xi

i. The disturbance term Ui is not correlated with explanatory variables. It means Ui’s &
Xi’s are not moving together or the covariance between Ui & Xi’s are zero
Cov (UiXi) =0
Cov (UiXi)
E [Ui − E (Ui)][ Xi − E ( Xi)]
By assumption we have E(Ui)=0 then
= E{[Ui-0][Xi-E(Xi)]}
= E{UiXi-UiE(Xi)]
= E(UiXi)-E(Ui)E(Xi)]
Again by assumption E(Ui)=0
= E(UiXi)-0E(Xi)]
= E(UiXi) since the value of Xi's are fixed then
=XiE(Ui)=0
Cov (UiUj)=0
ii. The explanatory variables Xi's are measured without error i.e no problem of
aggregation, round off etc. If there is such problem in the measurement it will be
absorbed by the random term Ui.

2.3.3. Relationship about explanatory variables

If there are more than one explanatory variable the relationships is assumed that they are
not perfectly correlated with each other.

Haramaya University 20 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Ex.
Yt=  + 1 X 1 +  2 X 2 +  3 X 3 + Ui
X1&X2, X2&X3, X1&X3 are not correlated with each other. i.e. no multicollinearity.

The distribution of the dependant variable Y

Given the following relationship between variables

Yi =  + Xi + Ui
Mean of Yi (Expected value of Yi) can be found as follow

E(Yi)= E[  + Xi + Ui ]
E(Yi) =  + Xi + E(Ui)
Where E (Ui) = 0 by assumption
E(Y) =  + Xi ----- is the mean value of the dependent variable Yi
Variance of Yi
Var (Yi) = E [Yi-E (Yi)]2
Substitute in place of E(Y) =  + Xi
Var (Yi) =E [Yi-E (  + X ) ]2
Again in place of Yi substitute Yi =  + Xi + Ui
Var (Yi) = E   + Xi + Ui −  − X i 2
Var (Yi)= E(Ui)2

From our previous assumption the variance of Ui is equal to E (Ui)2 =u2 then
Var (Yi)=E(Ui)2 = u2 which is constant.
The distribution of Y with mean & variance will be
Yi_N (  + Xi,  u )
2

2.4. Estimation of the model

The relationship

Yi =  + Xi + Ui − − − − − − − − − − − − − − − −2.5
Holds for population of the values X&Y. Since these values of the population are
unknown we do not know the exact numerical values of  & β' s. To calculate or obtain
the numerical values of  & β we took sample observations for Y & X. By substituting
these values in the population regression we obtain sample regression which gives an
estimated value of  & β given by ˆ & ˆ respectively then the sample regression line is
given by

Yˆi = ˆ + ˆXi − − − − − − − − − − − − − − − −2.6

The true relationships between variables (that explain the population) is given by

Yˆi =  + Xi + Ui − − − − − − − − − − − − − − − − − 2.7

Haramaya University 21 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

If you estimated this relationship using sample observation we get the estimated
relationship which has the following

Yˆi = ˆ + ˆXi + Ui − − − − − − − − − − − − − − − −2.8

We can estimate the value of &β using least square method (OLS) or classical least
squares (CLS).The reasons to start or use OLS or CLS methods are many
i. The parameters obtained by this methods have some optimal properties i.e.
BLUE (Best, Linear, Unbiased Estimators).
ii. The computational procedure of OLS is fairly simple as compared to other
econometric methods.
iii. OLS is one of the most commonly employed methods in estimating
econometric models.
iv. The mechanics of OLS is simple to understand.
v. OLS is an essential component of most other econometric techniques
From the sample observations we will have

Yˆi = ˆ + ˆXi + ei − − − − − − − − − − − − − − − 2.9

ei = Yˆi − ˆ − ˆXi − − − − − − − − − − − − − − − −2.10
Finding values for the estimates ˆ & ˆ which will minimize the square of residuals
 ei 2
^ ^

 ei 2 = [Yi − ˆ − ˆxi]2 − − − − − − − − − − − − − 2.11

i =1 i =1

To find the values of  & β that minimize this sum, we have to differentiate with respect
to ˆ & ˆ & set the partial derivatives equal to zero.

  ei 2
= −2 (Yi − ˆ − ˆXi) = 0 − − − − − − − − − − − −2.12
ˆ
  ei 2
= −2 (Yi − ˆ − ˆXi) Xi = 0 − − − − − − − − − − − − − −2.13
ˆ
First take equation number 2.12 to find the value of ̂

− 2 (Yi − ˆ − ˆXi) = 0 Run the sum over the equation

− 2 (Yi −  (ˆ − ˆXi) = 0

ˆ = nˆ
− 2 Yi − 2nˆ − 2ˆ  Xi = 0

Haramaya University 22 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

− 2 Yi + 2nˆ + 2ˆ  Xi =0

2 Yi − 2ˆ  Xi = 2nˆ Divided by 2n to get ̂

ˆ =
 Yi − ˆ  Xi
n n
 Yi = Y
n
&
 Xi = x
n
ˆ = Y − ˆx − − − − − − − − − − − − − − − −2.14
Take equation number (2.13) to find the value of ˆ

− 2 (Yi − ˆ ) − (2ˆ  Xi) X = 0 Multiply by X

− 2 (YiXi − ˆXi −ˆXi 2 ) = 0 Sum it over

− 2( YiXi − ˆ  Xi −ˆ  Xi 2 ) = 0

YiXi = ˆ  Xi + ˆ  Xi 2
− − − − − − − − − − − − − − − 2.15

Substitute equation (2.14) in to equation (2.15 )

YiXi = (Y − ˆx ) Xi + ˆ  Xi 2

YiXi = Y −  Xi − ˆx  Xi = ˆ  Xi 2

YXi − Y  Xi = −ˆx  Xi + ˆ  Xi 2

We know that Y =
 Y & x =  X substituted
n n
 Yi Xi
= − ˆ
 Xi Xi + ˆ Xi 2
 YiXi − n n
 
 Yi Xi ˆ ( Xi) 2 ˆ
YiXi − n = − n +   Xi 2 Multiplied both sides by n
n YiXi −  Yi Xi = − ˆ ( Xi) 2 + ˆ  Xi 2 n
n YiXi − Yi Xi = nˆ Xi 2 − ˆ
    Xi )
2
(
n YiXi −  Yi Xi = ˆ (n Xi 2
− ( Xi) 2 )
Haramaya University 23 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

n YiXi −  Yi Xi
ˆ = − − − − − − − − − − − − − − − −2.16
n Xi 2 − ( X ) 2
The numerical value of ˆ & ˆ can be found in deviation forms. To write the above
equation number 2.16 in deviation form
Take the numerator which is

n XiYi − Yi Xi
Added & subtracted Yi Xi
n XiYi − Yi Xi = n XiYi − Yi X + ( XiYi XiYi)
= n XiYi − Yi Xi +  XiYi −  XiYi
= n XiYi − Yi Xi −  XiYi +  XiYi

= n Xi  Yi −
 Yi Xi −n  Xi Yi + n  X n  Y
n
 n
 n n
Take n in common

n  XiYi − Y  X − x Y + nxy 
n ( Xi − x )(Yi − y ) -------------------------2.17

This equation is equal to the numerator of equation number 2.16.

Again from equation 2.16 take the denominator
n Xi 2 − ( Xi ) = n Xi 2 − 2  Xi + ( Xi )
2 2 2
( )
= n Xi 2 − 2 Xi  Xi + ( Xi )
2

= n Xi 2 − 2nx  X + n 2 x 2
= n( Xi − 2x  X + nx
2 2

= n ( xi − x ) − − − − − − − − − − − − − − − − − 2.18
2 2

By taking equation 2.17 as numerator & 2.18 as denominator

ˆ =  1
n ( X − X )(Y1 − Y )
n X 1 − X ) 2
X1 − X = xi & Yi − Y = yi Substitute in the above equation
n xyi
ˆ =
n xi 2

ˆ =
 xiyi = − − − − − − − − − − − − − − − − −2.19
 xi 2

Haramaya University 24 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

2.5. Statistical tests of Estimates

The two most commonly used tests in econometrics are r2 i.e. square of correlation
coefficient & the standard error of tests ( s.e.).

The Square of Correlation Coefficient = r2 /R2/

When we estimate a model of two variable case (one independent (X) & one dependent
variable Y) we find r2. But if we have more than two variable case (one dependent
variable & more than one independent variables (X1,X2...Xn) we will have the coefficient
of determination R2.
Definition of r2/R2/After estimation of ˆ & ˆ from the sample data observations of Y &
X using OLS method, we need to know how 'good' is this fit of the line to the sample
observations of Y&X. Means measure the dispersion of the sample observation around
the regression line. The closer the observations to the line the better is the explanation of
the variation of Y by the change in the explanatory variables (X's)
r2 shows the percentage of the total variation of the dependent variable that can be
explained by the independent variable X.

Y
Yi = ˆ + ˆxi
Unexplained variation X

Total variation

Y
Explained variation

Suppose a researcher may have Yi=+βXi+ Ui model. To estimate this model he took
some sample observation to estimate the value of  & β. In his estimation all the data
may fall below, above or on the line. Then using R2 he can observe that whether the
regressions line will give the best fit for these data or not.
• Yi is the observed sample value
• Y is the mean value of the sample

Haramaya University 25 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

• Ŷ is the estimated regression line using sample data

• Yi - Y = shows by how much the actual sample value is deviating from the
sample mean value. This is called total variation represent by small y.
• Yˆi − Y Explain by how much the estimated values are deviating from the sample
mean value. This is called explained variation & represent by ŷi
• Y − Yˆi This also shows that the difference between the actual value of Yi & the
estimated value of Yi ( Yˆi ). This is called unexplained variation represent by ei.
Therefore;
Total variation: yi = Yˆi − Y − − − − − − − − − − − − − − − −2.20
Explained variation yˆ = Yˆ − Y − − − − − − − − − − − − − − − 2.21
Unexplained variation ei = Yi − Yˆi − − − − − − − − − − − − − − − 2.22
Sum it over each equation & squared it. We will have

 yi
=  (Yi − Y ) 2 − − − − − − − − − − − − − − − −2.23
2

We square it because the sum of deviation of any variable around its mean value is zero
then to avoid this we make it squared.

 yˆi =  (Yˆi − Y )
2 2
− − − − − − − − − − − − − − − − − 2.24
 ei =  (Yi − Yˆ )
2 2
− − − − − − − − − − − − − − − − − −2.25
We can write equation no (2.20) as follows

yi = Yˆi − Y  Yi = yi + Y − − − − − − − − − − − − − − − −2.26
From equation (2.21)
yˆi = Yˆ − Y  Yˆi = yˆi + Y − − − − − − − − − − − 2.27
Substitute these equations in equation number 2.21
ei = Yi- Ŷ from the above equation (2.26 &2.27 ).
Yi = yi + Y
&
Yˆ = yˆ + Y
( ) (
ei = yi + Y − yˆi − Y )
ei = yi + Y − yˆ − Y
ei = yi − yˆi
yi = ei + yˆi − − − − − − − − − − − − − − − 2.28
This shows that each deviation of the sample observed values of Y from its mean
Yi − Yˆ = yi consists of two components
yˆi = Yˆi − Y which shows the explained amount by the regression line
ei + Yi − Yˆ = the unexplained variation by the regression line
By Taking equation number 2.28 yi = yˆi + ei Sum it over

Haramaya University 26 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 yi =  ( yˆi + ei) Squared it

^ ^

 yi
i =1
2
=  ( yˆ i + ei) 2
i =1

 yi 2
=  yˆi 2 + 2 yˆei +  ei 2 − − − − − − − − − − − − − − − − − 2.29

From this equation 2  yˆi ei is equal to zero. We can prove it.

yˆ = Yˆ − Y We know that from equation 2.14

Y =  + ˆxi again
Y = ˆ + ˆx Then if we substitute these in yi = y − Y
yi = ˆ + ˆXi − ˆ − ˆx
yi = ˆ ( Xi − x )
( Xi − x ) = xi is in deviation from
yi = ˆxi − − − − − − − − − − − −2.30
We also know that ei = yi − yˆi since yi = ̂xi
Substitute ei = yi − ˆxyi − − − − − − − − − − − − − − − 2.31
= 2 yˆiei = from − equation − 2.29 ˆxi( yi − ˆxi)
= ˆ xi( yi − ˆxi)

= ˆ  xyi − ˆ  xi
2 2

From equation 2.19 we know that ˆ =

 xiyi then substitute in the above equation in
 xi 2

place of
ˆ
 yˆei = ˆ  xiyi − ˆ  xi
2 2 2

= ˆ  xyi − ˆ  xi  2

 xiyi  xiyi −  xiyi xi 

=   2

 xi  2
 xi 
2

 xiyi
 ŷiei = xi  xiyi −  yixi = 0
 2

 ŷiei = 0 Then equation number 2.29 can be written as follow

 yi =
2
 yi +  ei 2
− − − − − − − − − − − − − − − − − −2.32
2

Total variation = Explained variation + Unexplained variation

Divided both sides of equation number (2.32) by  yi 2 & you will get

Haramaya University 27 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 yi =  yˆ +  ei
2 2 2

 yi  yi  yi
2 2 2

1=
 yˆ +  ei
2 2

 yi  yi2 2

 yˆ 2

Is the % of explained sum of squares which is r2

 yi 2

1= r +
 ei − − − − − − − − − − − − − − − 2.33
2
2

 yi 2

Then r =
 yˆi − − − − − − − − − − − − − − − 2.34
2
2

 yi 2

Again from equation 2.32

r 2
= 1−
 ei 2

− − − − − − − − − − − − − − − 2.35
 yi 2

ˆ
2.6. The relationship between r2 and the slope coefficient  .
We know that

r2 =
 yˆi 2

.
 yi 2

From equation 2.21

yˆ i = Yˆ − Y
yˆi = ˆ + ˆXi − ˆ − ˆx
yˆi = ˆ ( xi − x )
yˆi = ̂xi Substitute in equation 2.34
From equation
 yˆ 2  ( ˆxi 2 ) ˆ  xi 2
2

r =
2
= =
 yi 2
 yi 2
 yi 2
From equation number 2.19 we know that

ˆ =
 xiyi
 xi 2
Then substitute in place of ˆ

ˆ  2 .  2
2
xiyi xi
 xi  yi
r 2 = ˆ
 xyi .
 yi 2

Haramaya University 28 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Or if you further substitute ˆ again =

 xiyi
 xi2
r 2= ˆ
=
 xiyi  xiyi
. =
( xiyi) 2
− − − − − − − − − − − −2.36
 xi 2  yi 2  xi 2 . yi 2
Then the value of r2 can be found in different ways

r 2=
 yˆi 2

. or 1−
e 2

or 
 xiyi or ̂  xi 2

or
( xiyi) 2
 yi 2
 yi 2
 yi  yi2 2
 xi . yi
2 2

Limiting values of r2

r 2
= 1−
 ei 2

 yi 2

We have three options

1. If all the sample observations lie on the regression line there will be no scatter point’s
i.e the difference between the actual value & observed value is zero i.e.
ei = Yi- Y =zero. Then r2 will be 1
2. If the regression line explains a part of the actual values i.e. some are not explained

by the regression line then ei will have some values

 ei 2
will be greater than zero
 yi 2
but less than one then the r2 will lie between zero & one 0  r 2  1

3. If the regression line does not explain any part of the variation then
 ei 2

will be
 yi 2

one then r 2 will be zero.

2.7. Significance test of the Parameter estimates

The value of ˆ & ˆ which are obtained from the sample observation must be tested
whether they are statistically reliable to explain the population parameters or not. To test
the statistical reliability of the sample estimates we should know their mean & variance.
To do this we follow the following steps.
a) develop the formula for the computation of the mean & variance of the least
square estimates
b) Explain the statistical significance of the estimates using standard error & t-test
c) Construct confidence intervals for the estimates ˆ & ˆ .
Mean & Variance of the least square estimates
• Mean of ˆ
We can obtain mean value of ˆ from equation number 2.20

Haramaya University 29 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

ˆ =  xiyi .
 xi 2

In place of yi substitute equation number 2.21 then

ˆ = 
xi(Yi − Y )
.
 xi 2
=
 xiYi − Y  xi .
 xi 2

By definition it is known that the sum of any variable deviations from its mean is equal to
zero. Then  xi = 0 then Y  xi = 0

ˆ =
 xiYi . =  xiYi − − − − − − − − − − − − − − − − − − − 2.37
  xi 
 xi 2 2

The value of the independent variables is a set of fixed values, which do not change from
xi
sample to sample. Then will be a constant number & lets’ represent it by K &
 xi2
equation 2.37 can be written as
ˆ =  kiYi − − − − − − − − − − − − − − − − − − − − − − − 2.38
We know that from equation number 2.5

Yi =  + i + ui Substitute in equation 2.38

ˆ =  ki( + Xi + ui)
ˆ =   ki +   kiXi +  kiui − − − − − − − − − − − − − 2.39
 ki = 0 We can prove it as follows
ki =
xi
=  ki =
 xi By definition
 xi 2
 xi 2

 x = 0 Because the sum of any variables deviations from its mean is zero
Then
 xi =  ki = 0
 xi 2

Again  kiXi = 1

To prove it we know that Ki =

 xi
 xi 2

 xiX
 KiXi = xi − − − − − − − − − − − − − − − − − −2.40
 2

xi = ( Xi − x ) Substitute in the above equation 2.40

Haramaya University 30 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 ( X − x)x
 kiXi =
 xi 2

=
 X − x  x − − − − − − − − − − − − − − − − − − − − − 2.41
2

 xi 2

We can further simplify this as follows. Take the denominator

 xi 2 =  ( X − x )2 =  X 2 − 2xX + x 2  
= X 2
− 2 x  Xi + nx 2
Since x 2 is constant number summation of xi means multiply by n. Again here x means
 X then substitute in place of x .
n
 X . X + n  X 
2

X 2
−2
n
  n 
 
( X ) 2 ( X ) 2

X2 −2 n
+ n.
n2
( X ) 2 ( X )2
X 2
−2
n
+
n2
 X  X
 X 2 − 2 n  X  + n X
 
 X − 2x  X + x  X
2

x =X
2 2
− x  X substitute in the denominator of equation number 2.41
X 2
− x X
 KiXi = X =1
 2
− x X
Then  kiXi = 1
xi = Xi − x
Xi = xi + x
Substitute in the equation  kiXi in place of Xi
xi
 ki( xi + x ) We know that ki =
 xi2
=
 xi( xi + x ) =  xi + x  xi 2

 xi 2
 xi 2

 xi +  xi again by definition xi = 0
2

= 
 xi  xi
2 2

 xi
2

 kiXi = xi = 1
 2

Haramaya University 31 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Now we turn in to equation number 2.39 & write it as follows (Remember

that  xi = 0 &  kiXi = 1)
ˆ =  +  kiUi

 xi substitute it
Again  ki =
 xi 2

ˆ =  +
 xiui
 xi 2

Take the expected (average or mean value of ˆ )

E ( ˆ ) = E (  ) + E (
 xiui) Since xi’s are constant
 xi
= E ( ) +
 xiE ui
 xi 2

Again from the assumption number 2 of OLS E (ui)=0

E ( ˆ ) =  - - - - - - - - - - - - - - - - - - - - - - - - - - - - -2 . 42
The mean of OLS estimate ˆ is equal to the true population parameter mean value i.e 
Variance of ˆ
Var ( ˆ ) + E[ ˆ -  ( ˆ )]2
From equation 2.42 we know that  ( ˆ ) =  substitute it
Var ( ˆ ) = E[ ˆ -  ] 2
From equation 2-38 we know that

ˆ  xiyi  xi(Yi − y )
= = =  kiYi
 xi 2  xi 2
Then Var ( ˆ ) = var (  kiYi ) =  ki 2 var ( Yi ) - - - - - - - - - - - - - - -- - - 2.43
xi
ki =
 xi 2
First let’s find var (Yi) = E[Yi − E(Yi)]2
We know that Yi =  + Xi + Ui
E (Yi) = E ( ) + E ( Xi) + E (ui)
E (Yi) =  + xi + 0
By assumption number 2 E(ui) = 0
Substitute  + Xi + Ui in place of Y &  + Xi in place of E(Yi). then
2
Var (Yi) = E[ + Xi + Ui − ( + Xi)]
= E[ + Xi + Ui −  − Xi]2
Var Yi = E[Ui]2 = u 2 - - - - - - - - - - - - - - ----------- - - - -2.44
From equation no 2.43 we have the following

Haramaya University 32 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Var ( ˆ ) =  ki 2 Var ( Yi )
From equation no 2.44 Var (Yi) = u 2
Var ( ˆ ) =  ki 2u 2
2
 xi 
= u 2  ki 2 - - - - - We know that  ki 2 = E  2
  xi 
2
 xi 
= u  
2
2
  xi 
 xi 2
= u
( xi 2 ) 2

Var ( ( ˆ ) =  2
1
- - - - - - - - - - - - - - -- - - - - - -2.45
 xi 2
 2
Now we can say that ˆ has a mean of  & Variance of
 xi 2
u 
( , 
Then ˆ ~ N  xi 
Mean of ̂
 xiyi
From Equation no 2.14 we have ˆ = yˆ − ˆxˆ a gain from equation 2.19 ˆ = &
 xi 2
from equation number 2.38 we have ̂ =  kiyi substituted
ˆ = y − ( kiYi) x
= y − x −  kiYi
 Yi
= − x  kiYi
n
Take Yi &  as common b/c x & ki are constant
1 
̂ =  − x ki Yi - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - 2.45
n 
Take the expected Value of ̂ since n, x & ki are constant the expected value of a
constant is a constant itself.
1 
E (ˆ ) =   − x ki  (Yi) We know that
n 
E (Yi) = E ( + xi + Ui)
=  + xi +  (i)
By definition E(ui) = 0 then it will be  + xi
1 
E (ˆ ) =   − x ki  ( + Xi)
n 

Haramaya University 33 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 1 
=   − x ki + X + Xix ki 
n n 
n  xi
= − x  ki +  − x i  kixi]
n n
=  − x  ki + x  kiXi
We know that  ki = 0 &  kixi = 1
=  + x − x
E (ˆ ) =  - - - - - - - - - - - - - - - - - - - -- - - - - - -- -- - 2.46

Variance of ̂
1 
Var (ˆ ) = E[ − E(ˆ )]2 we know that from equation no 2.45 that ˆ =   − x ki  yi then
n 
substitute it
 1 
Var (ˆ ) = Var  E[ − x ki]Yi
 n 
Since n, x & ki are constant number their variance is constant.
2
1 
Var (ˆ =   − x ku  var Yii
n 
We know that Var Yi= u from equation no 2.43 substitute in place of Var Yi. Then we
2

will have.
2
1 
Var (ˆ ) = u   − x ki 
2

n 
1 1 
= u 2   2 + x 2 ki 2 − 2 x ki 
n n 
We know that the summation over a constant number is equal to multiplying the constant
number by n
n 2 x  ki 
= u  2 + x 2  ki 2 −
n n 
We proved that  ki = 0
2
  xi   xi 2 1
Again  ki = 
2
 = =
  xi  ( xi  xi 2
2 2

Substitute these values

n 1 − 2 x (0) 
= u 2  2 + x 2 −
n  xi 2
n 
1 x2 
= u 2  + 2 
If you take the common denominator n  xi 2
 n  xi 
2   xi x n 
2 2
ˆ
Var ( ) = u   
2 
 n  xi 

Haramaya University 34 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

We proved that  xi 2 = ( Xi − x ) 2 =  xi 2 − nx 2 (you can refer the devation from

equation no 2.40)

We can summarize the mean & variance of the variables as follows


Yi ~ N ( + xi),u 2 
   xi 2 + x 2 n 
ˆ ~ N  , u 2  
 n  xi
2
 
 u 2 
ˆ ~ N   , 
  xi 2 
Standard Error (S.E) values of ˆ & ˆ

S.E is the square root of variance

S.E ( ˆ ) = var(ˆ ) = ˆ
2
− − − − − − − − − − − − − (2.47)
 xi 2

S.E ( ˆ ) = var(ˆ ) = ˆ 2
−
X 2

− − − − − − − − − − − − − (2.48)
n xi 2

Sine ̂ cannot be easily computed it will be substituted by

2  ei 2

Equation 2.47 & 2.48

n−2
can be written as follows

ˆ ) =
 ei 1
2

.
S.E (
n − 2  xi 2

S.E ( ˆ ) =
 ei . X
2 2

( n  xi )(n − 2)
2

Standard error test of the least square estimates

From the sample observation of Y&X we can estimate or obtain the value of ˆ & ˆ .
Since the observations are sample taken from the population then due to sampling errors
the estimates may not truly explain the population. Then it is necessary to apply test of
significance in order to

a) Measure the size of the error committed

b) Determine the degree of confidence that the estimates will lie.
Though there are other methods used for the purpose, we use standard error test which is
very popular in many econometric research. The standard error test (S.E) will help us to

Haramaya University 35 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

determine whether the estimates ˆ & ˆ are coming from a population whose parameters
are zero called the null hypothesis.
Ho =  i=0
This means there is no relationship between the dependent & independent variables.
Or the samples from which ˆ & ˆ are coming from a population whose parameters are
different from zero are known as alternative hypothesis
H1 = i  0
This means there is a relationship between the dependent & independent variables.

Interpretation of S.E
Once we obtain the standard deviations of ˆ & ˆ using equation number 2.47 & 2.48
respectively& by comparing these estimated values with their s.e we can interpret the
results as follows.

Economic interpretation
If we have Yi = ˆ + ˆxi
Case A) Acceptance of the null hypothesis means

Accept Ho =  i=0
ˆ
Reject the alternative that H1 = i  0. This will occur if S.E ( ˆ ) 
2
It means the estimation are statistically insignificant or
✓ We accept the null hypothesis that the true parameter  is zero or
✓ The independent variable is insignificant.
✓ The sample parameter do not explain the population parameter or
✓ The independent variable do not influence the dependent variables (no
relationship between the dependent & independent variables)
Then the above equation will be written as Yi = ̂

Because the value of Ho = ˆ is not different zero or the slope of the line is zero
Again in case of the intercept ̂
If we accept the null hypothesis HO=  =0 & reject the alternative H 1 =   0 .
ˆ
This will occur if S.E ( ˆ ) 
2
And the meaning of ̂ is statistically insignificant or the equation will not have intercept
or the intercept is not differed from zero. Then
Yi = ̂xi Since ˆ = 0

Haramaya University 36 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Case B) Reject the null hypothesis & accept the alternative

i. Reject Ho = ̂ 0 =0 & accept the alternative that H1 = 1  0 .
ˆ
This will occur if S.E ( ˆ ) 
2
It means
✓ The estimates are statistically significant
✓ The estimates are significantly different from zero
✓ The independent variables will influence the dependent variable or there is a
relationship between the dependent & independent variables.

ii.Reject HO= ̂ =0 & accept the alternative that H 1 = ˆ  0 . This will be observed if
ˆ
S.E( ˆ ) 
2
Again it means the equation will have intercept or ̂ is statistically significant

Geometric interpretation of the S.E. test

ˆ
1. If we find that if S.E ( ˆ )  we accept the null hypothesis that HO= ̂ =0 & reject
2
the alternative H 1 = ˆ  0 . In this case if we have the equation Yi = ˆ + ˆxi since
ˆ = 0 we will write this equation as Yi = ̂xi and we do not have the intercept. Then
the equation will pass through the origin (fig. a)

Yi = ̂xi Yi = ˆ + ˆxi

X
0 0
Fig (a) fig. (b)
ˆ
2. If we find S.E. ( ˆ )  we will reject the null hypothesis that HO= ̂ =0 & accept the
2
alternative H 1 = ˆ  0 . Then in this case we will have intercept term & the equation
will be Yi = ˆ + ˆxi (fig. b)
ˆ
3. If we find that S.E ( ˆ )  we accept the null hypothesis that Ho = ̂0 =0 & reject
2
the alternative that H1 = 1  0 . Then the equation will be Yi = ̂ because ˆ = 0 set

Haramaya University 37 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Yˆ = ˆ + ˆxi
Y = ̂i

0 X 0
Fig. c Fig. d X
ˆ
4. If we find that S.E ( ˆ )  we reject the null hypothesis Ho = ̂ =0 & accept the
2
alternative that H = ˆ  0 . Then the equation will be Yˆ = ˆ + ˆxi
1 1

The student t- test

The second statistical test, next to S.E. will be under taken using t-test. This t-test will be
applicable if the sample size is less than 30 and the population parameters distribution is
normal. To apply t-test we follow the following steps.
i. Define the null hypothesis & alternative hypothesis
The null hypothesis Ho = ̂i =0 Alternative hypothesis = H = ˆ  0 . 1 1

ii. Choose the level of significance (5% or 1%)

Level of significance shows that rejecting a null hypothesis when it is true [committing
type I error). Or choosing a certain level of probability with which we would be willing
to risk error Type I is called significance level.
Ex. If the level of significance is 5%, then there are 5 chances out of 100 that we would
reject the null hypothesis (i.e. you commit error & reject Ho = ˆ =0 when it is correct &
i

accepted)
OR we are 95% confident that we have made the right decision; & only with 5% of
probability that we might have done wrong.
iii. Define the number of degrees of freedom (d.f.). i.e. N-K
N = total number of sample size
K = number of estimated variables (ˆ & ˆ )
The d.f = N-K
ˆ
t=
S .E ( ˆ )
ˆ
a) If t  we reject the null hypothesis Ho = ̂0 =0 & accept the alternatives,
S .E ( ˆ )
i.e. H1 = ˆ1  0 .
OR t= falls in the critical region

Haramaya University 38 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

ˆ
b) If t  we accept the null hypothesis Ho= ̂0 =0 & reject the
S .E ( ˆ )
alternatives H1 = ˆ1  0 .
OR t= falls in the acceptance region

Ho = ̂0 =0
H1 = ˆ1  0 . Acceptance region
H1 = ˆ1  0 .
Rejection region
Rejection region

-2.228 0 + 2.228
Figure (e)

Interpretation of t test
ˆ
If t  it means reject the null hypothesis Ho = ̂0 =0
S .E ( ˆ )
Accept t the alternative H = ˆ  01 1

✓ The estimated values are statistically significant or

✓ The sample observation can explain the population parameters.
✓ There is a relation between the dependent & independent variables.
Then the t-value will lie in the critical region (shaded) Suppose if we chose 5% level of
significance and the total number of sample size is 12 & the number of estimated
~
variables are ( ˆ &  ) only two. Then the d.f will be 12.-2=10. Again if we are
 5% 
undertaking two tall test then we will have 0.025   in each side (see figure e)
 2 
Then to find the t-value at 5% (0.25 each side) & 10 d.f. We can read from the t-table as
follows
a) In the first raw find 0.025
b) In the first column find 10 then at the intersection point of the raw & the column
you will get t-table value 2.228.
ˆ
Again if t  it means accept Ho = ̂0 =0
S .E ( ˆ )
Reject the alternative hypothesis H = ˆ  0 1 1
✓ The estimated values are statistically insignificant

Haramaya University 39 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

✓ No relationship between the dependent & independent variables

✓ The sample observations do not explain the population parameters.
Then the t-table value will lie in the acceptance region

✓ Simple inspection to calculate t-test

At 5% level of significance we can follow the following rules. Calculate t-values by

ˆ
. Then if this value is greater than +2 or less than -2 we reject the null hypothesis
S .E ( ˆ )
Ho = ̂ =0 & accept the alternative H = ˆ  0 . And the t-value will lie in the critical
0 1 1

ˆ
region. If calculated t = smaller than +2 & greater than -2 or (if -2<t<2) we
S .E ( ˆ )
accept the null hypothesis & reject the alternative.

✓ Confidence intervals for ˆ & ˆ

When we accept the alternative hypothesis that H 1 = ˆ1  0 & reject the null hypothesis
that Ho = ̂ 0 =0. It doesn’t mean that our estimates ˆ is the correct estimate of the
population parameter ˆ . It simply means that our estimates come from a population
whose parameters are different from zero (there is a relationship between the dependent
and independent variables). Then by constructing confidence intervals we can define how
our estimates are closer to the true population parameters. Interpretation of confidence
intervals. If we have 95% confidence level.
a) In the long run 95 out of 100 cases will contain the true parameter i in the
limit or intervals. OR
b) We are 95% confident that the unknown population parameters ( i ) will lie
within the limit /interval. OR
c) In 5% of the cases the population parameter will lie outside the confidence
limits.
But it doesn’t mean that or we cannot say that the confidence interval contains the true
population parameter ( i ). Because the probability contains specific fixed interval is
either 1 or 0. Confidence interval from the student t- distribution

ˆi − 
t= − − − − − − − − − − − − − − − − − − − − − − − − − − − −2.48
S ( ˆ )
With in (n-k) d.f.
ˆ = estimated value from the sample
 = is the population parameter

Haramaya University 40 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

If we choose any confidence level say 95%. We find from the t-table i.e the value
t + 0.025 with d.f of (n-k).
P { t -0.025<t<t+0.025} = 0.95-------------------------------------------2.49
Substitute equation number 2.48 in equation number 2.49
ˆi − 
P{-t0.025 <t0.025} = 0.95
S .E ( ˆ )
Multiply both sides by S.E. ( ˆ )
P{-t0.025 (S.E. ( ˆ ) ) ˆ −  <t0.025.S.E. ( ˆ ) } = 0.95
i = ˆ  t 0.025S .E.(ˆ ) − − − − − − − − − − − − − − − − 2.50
This confidence interval shows that the unknown population parameter i will lie with in
the defined limit 95 times out of 100. OR we are 95% confident that the unknown
population parameter i will lie with in this limit.
Ex.1.
In Table 2.1 the investment expenditure and the long run interest rate over the ten year
period is given. Test the hypothesis that investment is interest elastic by fitting regression
line to the data given & conducting the relevant test of significance. To answer the
question we follow the steps of econometric methodology.
i. The economic theory states that “investment is interest sensitive.”
ii. Mathematical model of the theory
a. Selecting variables:- Investment is the dependent variable (endogenous)
and interest rate is exogenously determined (independent variable)
b. A priori expectation of the sign of the parameters:- There is a negative
(inverse relationship) between Investment & rate of interest
c. Magnitude of the parameter:- The theory says that investment is interest
elastic then the coefficient is greater than one.
Specification of the model:- This will be a single equation model which investment will
be affected by rate of interest but investment will not affect rate of interest. The
relationship between investment and rate of interest is assumed to be linear (This will be
determined by the researcher).
Then
Yt =  + Xt
Where Yt= is the level of interest
Xt = is rate of interest,  >1 &  is the intercept
iii. Specification of the Econometric model.
There are other variables which will affect investment these are marginal efficiency of
capital (MEC), saving, consumption, political stability etc. Since these & other variables
are not incorporated in the mathematical model of investment function we can capture
these variables by incorporating a random term Ui in our model.

Haramaya University 41 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Yt =  + Xt +Ui
Then by adding the random (error, or stochastic or disturbance) term Ui we convert the
mathematical (exact) relationship between investment & rate of interesting in to in exact
relationship of Econometric models.
iv. Obtaining data:- A sample of 10 years observation data are given to estimate
the model & the type of data are time series data.
v. Estimation of the econometric model
The economic relationship is explained using a single equation model then the most
appropriate method of estimating this equation is OLS method. To estimate this model
we use table 2.1 & obtaining the following results in deviation forms.
 yixi = −1.42  xi 2 = 0.00064
 yi = 25826.4
2
ei 2
= 25,654
 yi = Y = 753.4  xi = x = 0.056
n n

From equation number 2.18 we have ˆ =

 xiyi then
 xi 2

− 1.424
ˆ = = −2225
0.00064
ˆ = Yˆ − ˆxˆ From equation number 2.14
ˆ = 753.4 − (−1.424 x0.056) = 878
Then the estimated regression line will be
Yˆ = 878 − 2225 Xi
vi. Evaluation of the estimates:-
After estimation & obtaining values of the coefficient ( ˆ & ˆ ) of the variable we should
have to evaluate the results obtained using Econometrics method
a. Economic interpretation of the results:- at this stage check whether the obtained
results are economically meaningful or not.
Yˆ = 878 − 2225xi
i. If interest rate is zero i.e Xt =0 then Yˆ = 878 means if the interest rate is zero
investment will be equal to 878 birr. This is the interpretation of the constant
term ̂ in case of linear equation
ii. ̂i Indicates that if rate of interest increase/decrease by 1 birr then investment
will decrease/increase on the average by 2225 birr. Therefore it passes the
economic criterion because it explains the inverse relationship between
investment & interest rate.
b. Since the model passes a prior economic criterion, the next step is to test the
reliability of the estimated parameters using statistical tests using r2, & S.E tests.
i. the correlation coefficient test r2
To estimate r2 we can use the formula of 2.31-2.35 let’s us

Haramaya University 42 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

r 2 = ˆ
 xiyi
 yi 2
We know that ˆ = -2.225  xiyi = −5.424.  yi2 = 25,826.4
− 2.225(−1.424)
r2 = = 0.123
25.8264.4
This means 12.3% of the change investment is accounted (explained) by interest rate &
the remaining 87.71, is not explained by rate of interest but by some other factors
represented by Ui in our model.
ii. Standard error test:-

ˆ ) = ˆu
x 2
2

n xi 2

since ˆu cannot be obtained we can approximately

2
S.E (

measure using
 ei 2

n−2

S.E( ˆ ) =
 ei 2  xi 2

n−2 n xi 2

From our estimation we obtain the following values  ei 2

= 22657
 Xi 2
= .032 n-2=10-2=8  xi 2
= 0.00064
(22658) (0.32) 90.632
S.E( ˆ ) = = = 17701.563 = 133.046
(8) 8(0.00064) 0.00512
Again S.E ( ( ˆ ) can be obtained using equation number 2.47

S.E ( ( ˆ ) =
 ei 2

=
22658
=
22658
=2103.661
n−2  xi 2
8(0.00064) 0.00512
Having obtained the value of S.E (ˆ ) and S.E ( ( ˆ ) we can undertake the S.E Test using
hypothesis testing.
ˆ
Test of S.E (ˆ ) : If  S .E (ˆ ) then we can reject the null hypothesis Ho= ˆ = 0 &
2
accept the alternative that H1= ˆ  0 .
ˆ = 878 & S .E (ˆ ) = 133.046
ˆ 876
= = 439> 133.046
2 2
ˆ
Then  S .E (ˆ ) we can conclude that rejecting the null hypothesis & accept the
2
alternative.

Test of S.E ( ˆ ) : If S .E ( ˆ )  we can accept the null hypothesis and reject alternative.
2

Haramaya University 43 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 2225
S .E ( ˆ ) - 2103.661 & = = 1112.5
2 2

Since S .E ( ˆ )  we accept the null hypothesis & reject the alternative
2
The student t- distributions
Given the 5% level of significance level we can have the following approximate
measurements of t-test
887
t= = 6.66 since t-value is greater than 2 we can reject the null hypothesis
133.046
& accept the alternative
ˆ
t=
S .E ( ˆ )
Again if t> + 2 we reject the null hypothesis & accept the alternative
− 2225
t= = −1.058 since the t value is greater than -2 then we accept the null
2,103.661
hypothesis & reject the alternative. To summarize we can write the results as follows
Yˆt = 887 - 2225xt
S.E (133.046) (2103.66)
T (6.66) (-1.058)
The equation passed the economic or a priori criterion but we found different results in
statistical test.
a) ̂ Satisfied the statistical test because
̂
S.E. (ˆ ) < & also t>2 we can say that we reject the null hypothesis H 0 = ˆ = 0 &
2
accept the alternative H1 = ˆ  0 . It means that the equation should have an intercept
term.
ˆ
b)  Does not satisfies the statistical test because
ˆ
S.E. ˆ > & which means acceptance of the null hypothesis that H 0 = ˆ = 0
2
✓ ˆ is statistically insignificant
✓ Investment (Y) is not affected by the change in the interest rate or investment is
not interest sensitive.
✓ No relationship between investment and interest rate because ˆ =0
And reject the alternative H = ˆ  0 means ˆ is not different from zero.
1

Again this is supported by the t-test. Since t= -1.058 & which is greater than -2 or t is
found between + 2 then we can conclude that we accept the null hypothesis H = ˆ = 0 0

and
✓ The estimates ˆ is statistically insignificant

Haramaya University 44 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

✓ ˆ is not different from zero

✓ No relationship between investment & interest rate.

Example 2

Given the data on table 2.2. fit the data and estimate the income elasticity.

Haramaya University 45 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Table 2.1
Invest- ( Xi − x )
Rates of interest (Yi − Y ) ei =
Year ment yi2 xi2 yixi Estimated Yˆi ei2
Yi
Xi Yi Xi yi - Yˆi
1958 656 0.05 -97.4 -0.006 9476.76 0.000036 0.5844 766.75 -110.75 1265.56
59 804 0.045 50.6 -0.011 2560.36 0.000121 -0.5566 777.875 26.125 682.5156
60 836 0.045 82.6 -0.011 6822.76 0.000121 -0.9086 777.875 58.125 3378.516
61 765 0.055 11.6 -0.001 134.56 0.000001 -0.0116 755.625 9.375 87.89
62 777 0.06 23.6 0.004 556.96 0.000016 0.0944 744.5 32.5 1056.25
63 711 0.06 -42.4 0.004 1797.76 0.000016 -0.1696 744.5 -33.5 1122.25
64 755 0.06 1.6 0.004 2.56 0.000016 0.0064 744.5 10.5 110.25
65 745 0.05 -6.4 -0.006 40.96 0.000036 0.0384 766.75 -10.75 390.0625
66 696 0.07 -57.4 0.14 3294.76 0.000196 -0.8036 722.25 -26.25 689.0625
67 787 0.065 33.6 0.009 1128.96 0.000081 0.3024 733.375 53.625 2875.641
 Yi  Xi  yi  xi  yi  xi  yixi  ei
 yˆi  ei
2 2 2 2

=7,534 =0.56
=0 =0 =25826.4 =0.00064 =-1.424 =7534 =0 =22658
Ŷ =753.4 X =0.056

Haramaya University Distance

46 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Table 2.2

log Ŷ Log Y-
Ye Consu Income Log (Logy) estimate log Ŷ =
2
ar mption Xt Log Yt LogXt Log yt Log xi Log y xi2 (logx) d (e) ei2
Yt
0
58 8 17 0.9 1.23 -0.317 -0.399 0.1004 0.1596 0.1286 0.9163 -0.0132 . 000176
0.0279
59 12 27 1.079 1.431 -141 -0.1986 0.0198 0.0394 7 1.069 0.0101 0.000103

0.0054 0.0032
60 15 36 1.176 1.556 -0.044 -0.0737 0.00193 3 3 1.164 0.0121 0.000146
0.00124 0.0010 0.0011 0.0103
61 18 46 1.255 1.663 0.0352 0.0327 4 7 5 1.245 7 0.000108

62 22 57 1.342 1.756 0.1224 0.1258 0.0149 0.0158 0.0154 1.316 0.0267 0.000716
-
0.0072
63 23 67 1.361 1.826 0.1417 0.196 0.02 0.0384 0.0278 1.369 9 0.0000531
-
0.0166
64 26 81 1.415 1.908 0.1949 0.278 0.038 0.0776 0.0543 1.431 8 0.000278
log xi 2
 log Yi
log X  log yi 2
 log Ŷ  ei 2

log Yi
=0.337 logx
=11.37 =0.1965 4 =0.256 =8.51 =0.00158
8.53 Y X 4
=1.22 =1.624

Haramaya University Distance

47 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

2.8. Exercise for chapter two

1. Given the following table about the relationship between corn production & fertilizer
utilization

Corn 40 44 46 48 52 58 60 68 74 80
Fertilizer 6 10 12 14 16 18 22 24 26 32

If the relationship between the production of corn & fertilizer is linear.

a. Estimate the parameters coefficients, standard errors & test the significance of the
parameters?
b. What variations of the corn production are explained by fertilizers?
c. Interoperate the results obtained from the linear relationship? Had the relationship was
nonlinear what would be the interpretation of the coefficients?
2. Suppose that Mr. A is estimating a consumption function and obtain the following results
Ŷ = 15 + 0.81xi .
t (3.1) (18.7)
Given that Y is consumption & X is disposable income. If r2 is 0.99 & n is 19
a) Test the significance of X statistically using t-ratios.
b) Determine the estimated standard error of the parameters
c) Construct a 95% confidence interval for the coefficient of disposable income.

Haramaya University 48 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Chapter 3: Properties of the least square Estimates

3.1. The Gauss-Markov Theorem
Given the assumptions of CLS model, the least square estimates obtained from a sample possess
some optimal properties. These are the least square estimators are linear & unbiased estimators.
These are also the best of linear unbiased estimators; it means the estimators have minimum
variance with in the classes of linear unbiased estimators. This theorem is sometime referred to
as the BLUE theorem. An estimator of OLS said to be BLUE (best linear unbiased estimator) if
the following holds true.
a) It is linear, if both the sample parameters are a linear function of the dependent variable.
ˆ & ˆ are a linear function of Yi
b) It is unbiased if the expected value of the sample parameters is equal to the true value of
the population parameter.
E (ˆ ) =  E ( ˆ ) = i
Where ˆ & ˆ are sample parameters, and ˆ & ˆ are the true population parameter & E means
expected or average values.
c) It has minimum variance i.e. if the variance of ˆ & ˆ has smallest variable as compared
to any other variance obtained from other econometric methods.
Gauss Markov Theorem: - Given the assumption of CLS model, the least square estimators
satisfies the properties of BLUE

1) The property of linearity

From equation 2.19 we have value of ˆ

ˆ =  xiyi And from equation number 2.38 we proved that

 xi 2

xi
̂ =  kiYi Where ki =
 xi2
From the above of X is constant in the sample & ki is constant then
̂ =  kiYi i.e. an estimate ̂ 2 is a linear function of Y's or ˆ is a linear function of the
values of the dependent variable.
By the same analogy equation 2.14 we have
ˆ = Y − ˆx
Substitute equation no 2.46 in place of ˆ
̂ = Y − x  kiYi
 Yi − x
̂ =
n
 kiYi
Haramaya University 49 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1
̂ =
n
Yi − x  kiYi
1 
̂ =   − x ki Yi
n 
1
We know that , x & ki are constant then ̂ is a linear function of the dependent variable Yi
n
Thus both ˆ & ˆ are expressed as linear function of the Y's.
2) Unbiasedness:- The bias of the estimation is defined as the difference between its
expected value & the true parameter.
Bias =E ( ˆ ) − 
If the estimation is unbiased its bias is zero
i.e. E ( ˆ ) =  we proved this in the previous page equation number 2.42
Again bias =E (ˆ ) −  if the estimation is unbiased its bias is zero E (ˆ ) =  Also we proved this
& you can refer equation number 2.46
3) The minimum variance property (best estimator)
The property of minimum variance is the main reason for the popularity of the OLS method.
Best in this sense means definitely superior. One should know that when we say OLS estimator
is best estimator it will have a minimum variance as compared to any estimators obtained using
other econometric methods such as 2SLS, 3SLS. Maximum Likely hood estimators etc.

3.2. Importance of the BLUE Properties:-

One may ask that why do we attach so much importance to the BLUE properties of the OLS
estimates. But the reasons are:
a) The property of linearity is desirable because it facilitates the computation of the
estimates.
b) The property of unbiasedness by itself is not important. If we have a very large number of
samples, the estimators of the parameters obtained from these samples will on the
average give the true value of  's
c) Best (minimum variance as compared to others). The least square variance property will
be desirable if it is combined with small bias because an estimate may have zero variance
and yet have an enormous bias. The importance of this property is apparent when we
want to apply the standard test of significance for ˆ & ˆ and to construct the confidence
intervals for estimates.

3.3. Maximum Likelihood Estimation

A concept frequently utilized in econometrics is that of maximum-likelihood. The basic
important concept of maximum likelihood is the fact that different statistical populations

Haramaya University 50 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

generate different samples. i.e. any one sample being scrutinized is more likely to have come
from some populations rather than from others.
Ex.1
If one where sampling coins tosses & a sample mean of 0.5 were obtained (representing half
heads or half tails), it would be clear that the most likely population from which the sample were
drawn would be a population with a true mean of 0.5
Ex.1.
Assuming that [X1,X2,X3,X4,X5,X6,X7,X8] are drawn from a normal population with a given
variance but unknown mean. Again assume that these sample observations are drawn from
distribution A or distribution B.
Probability

Distribution A Distribution B

x6 x2 x3 x4 x5 x8 x7 x1

Given the distribution A & B. then if the true population were B, then the probability that we
would have obtained the sample shown would be quite small. But if the true population were A
then the probability that we would have drawn the sample would be substantially large. Select
the observation from population A as the most likely to have yielded the observed data. We
define the maximum likelihood estimator of  as the value of ˆ which would most likely
generate the observed sample observations Y1, Y2, Y3 --- Yn. Then if Yi is normally distributed
& each of Y's is drawn independently then the maximum-likelihood estimation maximizes. P(Y1)
P(Y2) . . . P(Yn). Where each P represents a probability associated with the normal distribution.
P(Y1) P(Y2)--- P(Yn) is often referred to as the likelihood function. The likelihood function
depends up on not only on the sample values but also in the unknown parameters of the

Haramaya University 51 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

problems (ˆ & ˆ ) . In describing the likelihood function we often think of the unknown
parameters as varying while the Y's (dependent variables) are fixed. This seems reasonable
because finding the maximum likelihood estimation involves a search over alternative parameter
estimators which would be most likely to generate the given sample. For this reason the
likelihood function must be interpreted differently from the joint probability distribution. In the
latter case the Y's are allowed to vary & the underlying parameters are fixed & the reverse is true
in case of maximum likelihood. Now we are in a position to search for the maximum likelihood
estimators of the parameters of the two variable regression models.

Yi =  + xi + ui − − − − − − − − − − − − − − − − − − − − − − − 3.1
We know that Yi N ( + xi,  2 )
I.e Y is normally distributed with mean ( + xi) & variance  2
Assume that all the assumptions of least squares and further assume that the disturbance term has
normal distribution. Will the estimators ( +  ) different from the least square estimators? Will
such estimators possess the desirable properties? In our model Yi =  + Xi + Ui sample consists
'n' observations in Y and X. Then we will have a mean value of ( 1 , 1 , x1 ) , ( 2 ,  2 , x 2 ) ,
(3 , 3 , x3 ) ... ( n ,  n , xn ) but a common variance of  2 . Why we will have a different mean but
a constant variance? The reason is simple that a random variable X assumes different value with
a probability of density function of f(x) but fixed values of Yi. The joint probability density
function can be written as a product of n individual density function
( )(
F[Y1 , Y2 , Y3 ...Yn / 1 + 1 x1 ,  2 ] = f Y1 / 1 + 1 x1 ,  2 f Y2 /  2 +  2 x2 ,  2 )
f [Y3 /  3 +  3 x3 ,  2 ]
This probability distribution may be written as follows
(
F (Y ) = 2  2 ) 1
2
Exp 3.3
Where exp. denotes exponential function. Then the likelihood function
L[Y1 , Y2 ...Yn ,  ,  ,  2 ] = P(Y1 )P(Y2 ) − − − P(Yn − 1) P(Yn )
Let the likelihood function is represented by L
 1  Y1 −  1 − 1 xi  
2
 1  Y2 −  2 −  2 x 2  2 
L = (2  )   (2 ) exp − 
2 − 2 2 − 2
1 1
ex −   
 2     2    

 1  Y3 −  3 −  31 x3  
2
 1  Y4 −  4 −  4 x 4  2 
( 2 )   (2 ) exp − 
2 − 2
−1 1
* 2 2 exp −    ---
 2     2    

Sum it over i.e. from 1 up to n.

−1

( )
2
L = 2 2
) n  exp

If you represent the N product factor by 

Haramaya University 52 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

−1

N (2 )
^ 2
L= 2
exp 3.6
i =1

The value of Y1,Y2 ...Yn are given but the value of  &  ,  2 are not known then this
function can be called likelihood function and denoted by Lf ( ,  &  2 )
1
Lf ( ,  , 2 ) = N (2  2 ) 2 exp 3.7
And the equation can be written as
 1  Yi −  − ixi  
2
1
Lf ( ,  , ) = N
2
exp −    − − − − − − -------------------3.8
2 2  2   
Take log of L
−n n
ln Lf = log 2 − log  2 − 1 2
2  Yi − i − xi − − − − − − − − − − − − − − − 3.9
2
2 2

Differentiate w.r.t.  ,  & 

2
setting these with equal to zero.
 ln Lf 1
= − 2  (Yi −  − Xi)(−1) = 0 − − − − − − − − − − − 3.10
 2
 ln Lf 1
= − 2  (Yi −  − Xi)2 Xi = 0 − − − − − − − − − − − − − −3.11
 2
 ln Lf n 1
 2
=− 2 +
2 2 4
 (Yi −  − Xi)2 Xi = 0 − − − − − − − − − − − − 3.12
Equation number 3.10 will be equal to
 (Yi −  − Xi)2 Xi = 0 − − − − − − − − − − − − − 3.13
Since − 1 2  2 = 0
Equation no 3.11 will be again equal to
 (Yi −  − Xi)( Xi) = 0 − − − − − − − − − − − − − − − − − 3.14
Since 1 2  2 = 0
Substitute equation 2.13 in to equation 3.14 and we will have
~
Yi = n~ +   xi − − − − − − -------------------------------------3.15
~ ~
YiXi = N  X +  Xi 2 − − − − − − − − − − − − − − − −3.16
The two equations again will give us the same normal equation OLS. From equation 3.12 we can
to obtain the value of  2
−n 1
+ = Yi −  − Xi) 2 =0
2 2
2 4

1
Since 4 = 0 we left with
2
−n
+  (Yi −  − Xi) 2 = 0
2 2

Haramaya University 53 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

n
 (Yi −  − Xi) = 2
2
2

  (Yi −  − Xi) = n
2 2

1
 =  (Yi −  − Xi)
2 2

n
(Yi − ˆ − ˆXi) 2
1
2 =
n

We know that Yˆ = ˆ − ˆXi
Then Yi- Yˆ − Ui or Yi =  + Xi + Ui
Yi = ˆ + ˆXi Then Yi- −Ŷ = Yi =  + Xi + Ui
Therefore
1
 2 =  u 2 − − − − − − − − − − − − − − − − − -3.17
n

The ML estimation is different from OLS estimator of OLS. The variance was
 ei 2

in OLS
n−2
1
but it is
n
 ui 2 in case of ML. Thus the variance of ML is biased estimator of  2 but it is
unbiased in the case of OLS. But as the sample size increases the ML variance converges to the
true population variance.

3.4. Exercise for chapter three

1) What are the properties that make the OLS estimators BLUE?
2) What are the important properties of BLUE as compared to other methods of Econometrics?

Haramaya University 54 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

3.5. Chapter 4: Multiple Linear Regression

4.1. Introduction
In a simple regression we study the relationship between the dependent variable (Yi) and only
one independent (explanatory) variable Xi. In this simple regression analysis, for example the
Quantity demanded (Y) is depend up on the price of the product alone other things being
constant (absorbed by Ui). But in a multiple regression analysis when a dependent variable (Yi)
is depends upon many explanatory variables (independent variables).

Ex. If quantity demanded is a function of price of the product, price of other goods (substitute &
complementary goods) income, wealth, previous year income consumption behavior etc. For the
sake of simplicity let’s consider a three variable case

Yi =  + 1 X 1 +  2 X 2 i + Ui − − − − − − − − − − − − − 4.1
Where Yi is dependent & X1 & X2 are independent (explanatory variables) & Ui is the stochastic
disturbance term. Alternatively we can write this equation as follows.

Yi = 1−12 + 11−2 X 1i + 12.−1 X 2 + Ui − − − − − − − − − 4.2

• 1.12 Measures the constant term & measuring the average value of Y when X1 & X2 are
zero.
• 11.−2 Measures the change in Y for a unit change in X1 alone (the effects of X1 on Y given
that X2 is constant.)
• 12−1 Measures the change in Y for a unit change in X2 alone (the effects of X2 on Y given
that X1 is constant.)
The value of 11.−2 & 12−1 are called partial repression coefficients or marginal coefficients but
not elastic ties & these value measures only the average value of Y because the regression line is
a linear regression line. But if the regression equation is nonlinear like the following Cob-
Douglas production function

Let Y= output, L is labor & K is capital and  1 , 1 &  2 are parameters then to estimate this
equation first transform in to linearity using log.
Log Yi= log 1 + 1 log L +  2 log k + Ui ------------------------4.4
Then here measures elastic ties not average values. In order to find the value of
 &  we use all OLS assumptions regarding the disturbance term Ui.

E(Ui) = 0, E(Ui) 2 =  2 R(Ui,Uj)=0

E(UiXi) = E(UiXj) = 0
In addition to this we assume further that there is no exact linear relationship between
explanatory variable X1 and X2 (no multicollinearity between X1 and X2)

Haramaya University 55 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Given n observations on Y1, X1 & X2 our problem is to estimate the value of ˆ1 , ˆ2 & ˆ . Then
we apply OLS to obtain their estimates ˆ , ˆ & ˆ . In the simple linear regression model we have
1 2
already defined that
ei = Yi − Yˆ − − − − − − − − − − − − − − − − − − − −4.5
Yi = ˆ − ˆ1 X 1 − ˆ 2 X 2
  ei 2 = (Yi − ˆ − ˆ X − ˆ X ) 2
 1 1 2 2

Differentiate w.r.t ˆ1 & ˆ & ˆ setting these values equal to zero.

  ei 2
= −2 Yi − ˆ − 1 X 1i − ˆ 2 X 2 i) = 0 − − − − −4.6
ˆ
  ei 2
= −2 X 1i[Yi − ˆ − ˆ1 X 1i − ˆ 2 X 2 i] = 0 − − − − −4.7
ˆ 1

  ei 2
= −2 X 2 i[ yi − ˆ − ˆ X 1i − ˆ 2 X 2 i] = 0 − − − − −4.8
ˆ
 2
From the above equations we obtain the followings normal equations

Yi = nˆ +   X i + ˆ  X i − − − − − − − − − − − − − − − −4.9

1 1 2 2

YiX i = ˆ  X i + ˆ  X i + ˆ  X i X i − − − − − − − − − 4.10
1 1 1̂ 1
2
2 2 1

YiX i = ˆ  X i + ˆ  X i X i + ˆ  X i − − − − − − − − − − − −4.11
2 1 1̂ 1 2 2 2

From equation number 4.9 we have

̂ = Y − ˆ1 x1 − ˆ2 x2 − − − − − − − − − − − − − − − − − − − − − − − 4.12
By substituting these values in ̂ in to equation 4.10 & 4.11 yields
YiX 2i =  X 1i(Y − 1 x1 −ˆ 2x2 ) + ˆ1̂  X 1i + ˆ2  X 2i X 1i − − − − − − − −4.13
From equation number 4.12 & 4.13 by rearranging we obtain
 ( X 1 i − x1 )(Yi − Y ) = ˆ1  ( X 1i − x1 ) 2 + ˆ 2  ( X 1i − x1 )) X 2 i − x2 )
(X i − x2 )(Yi − Y ) = ˆ1  ( X 1i − x1 ) ( X 3i − x3 ) + ˆ 2  ( X 2 i − x2 )
2
2

If you write the above equation in lower cases letters for deviations from means we can write the
above equations in the following ways.
 xiyi = ˆ  ix2 x i + ˆ  x ix i
1 1 2 1 2

 x2 iyi = ˆ2  x1ix2 i + ˆ2  x2 i

If you solve for ˆ & ˆ , we obtain
1 2

 x1iyi  x2 i −  x1ix2 i  x 2iyi

2

ˆ1 = - - - - - - - - - - - - - - -4.14
 x1 i  x2 i − ( x1ix2 i ) 2
2 2

 x 2 iyi  x1 i −  x x ix2 i  x1iyi

2

ˆ 2= - - - - - - - - - - - - - - ..4.15
 x x i  x2 i − ( x1ixx i ) 2
2 2

Haramaya University 56 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

4.2. Variance & standard errors of OLS estimators.

Having obtained the particular regression coefficients of (ˆ & ˆ ) we can derive the variance &
standard errors of these estimations in the manner as follows.

1
x  x + x  x − 2 x x  x x  u . . .. . . . . . .. 4.16
2 2 2 2

Var ( (ˆ ) =  + 1 2 2 1 1 2 1 2 2

 x  x − ( x x )
2 2
 n 
2
1 2 1 2

 x  2

Var ( ( ˆ ) = u  2
 . . . . . . . . . . . . . . . . . . . . . 4.17
2

  x  x − ( x x ) 
2 2 2
1 2 1 2

 x  2

Var ˆ = u 
2)
2
 . . . . . . . .. . . . . . . . . . .. . . . 4.18.
1

  x  x − ( x x ) 
2 2 2
1 2 1 2

Where u = 2  e1
2
, n is number of sample, k is number of parameters which are estimated .
n−k 1

Standard errors of ˆ & ˆ 's

S.E (ˆ ) = Var (ˆ ) . . . . . . . . . . .. . . . . .4.19

S.E ( ˆ1 ) = Var ( ˆ1 ) . . .. . . …. . . . . . . . 4.20
S.E ( ˆ ) = Var ( ˆ 2 ) . . . . . . . . .. . . . . . . 4.21

4.3. Test of significance of the parameter estimates of multiple regressions.

In the two variable regression case we have seen how to undertake the statistical test of
significance. In this multiple regression analysis or more than two variable cases we follow the
same procedures of statistical test of significance just like the two variable cases.

The Standard error test of significance. (S.E)

S.E test is the comparison of the numerical values of estimates ( ˆ , ˆ1 , ˆ 2 ) with their standard
errors. First we set the null hypothesis and the alternatives as follows.
The null hypothesis of ̂ Ho= ˆ = 0
The alternative hypothesis ˆ Ho= ˆi = 0
The null hypothesis means the intercept term (ˆ ) and the coefficients of the variables are
equal to zero.
Given yˆ i = ˆ + ˆ1 X i + ˆ2 X 2 i
ˆ & ˆ1 ˆ 2 are equal to zero i.e. if we the accept the above null
If the estimated values of 
hypothesis it means

Haramaya University 57 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

a) The dependent variable Yi is not explained (depend on) by the explanatory variables X1i
& X2i or there is no relationship between Y & Xi's .
OR
b) We accept the null hypothesis that there is no relationship between Yi & Xi's
OR
c) The estimates are not significantly different from zero or the estimates are insignificant.
The alternative hypothesis
H2= ˆ  0
ˆ
H1 =  i  0
It means
a) The value of estimators  ˆ & ˆ1 ˆ 2 is different from zero i.e there is a relationship
between Yi & X1 & X2 or Y is explained by X1 & X2
OR
b) The estimators  ˆ & ˆ1 ˆ 2 are significantly different from zero. ( they are not equal to
zero)
OR
c)  ˆ & ˆ1 , ˆ2 are statistically significant. Having these ideas in your mind we can
undertake the test as follows
The null hypothesis H0= ˆ = 0
Ho = ˆi = 0
The alternative hypothesis H1= ˆ 0
H1= ˆi  0
̂ ˆ
1. If we find that S.E ( ̂ ) > & S.E ( ˆ )'s > (if the estimated S.E is greater than half of
2 2
the estimators).
We accept the null hypothesis & interpret as stated above or
There is no relationship between Y & X1, X2
ˆ & ˆ1 ˆ 2 are insignificant ( X1 & X2 are not affected Yi )
ˆ & ˆ , ˆ are equal to zero
1 2

Reject the alternative hypothesis which says the  ˆ & ˆ1 , ˆ2 are different from zero
ˆ
2. If we find that S.E ( ˆ ) 
2
ˆ

& S.E ( ( ˆi ) 
2
It means if the S.E are less than half of the estimators then we can interpreter it as follows.
a) Accepting the alternative means the value of  ˆ & ˆ1 , ˆ2 are different from zero and
reject the null hypothesis which means the value of  ˆ & ˆ1 , ˆ2 is equal to zero.

Haramaya University 58 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

b) There is a relationship between the dependent variable Yi and the independent variables X1
& X2 or X1 & X2 explains Yi
c)  ˆ & ˆ1 , ˆ2 are significant
The student's t- test
In the t- test analysis first we should obtain t- values for each estimator & we compare it with the
theoretical or table values as follows.

A) Calculated value of t. ˆ Calculated t- value for the intercept term 

t=
S .E (ˆ )

ˆi Calculated t- values of the slope parameters

t=
S .E ( ˆ )

B) Theoretical or table value of t. This can be obtained from the table as follows
• First count the number of sample size & represent it by n.
• Second determine the level of significance represented by 
• Third determine whether you are taking one tail or two tail test
• Forth count the numb of estimated values i.e  ˆ & ˆ1 , ˆ2 & represent by k
• Then having determine it we can write the table value of t as follow
 
t   , (N-K)
2
 
  Shows two tail test
2
N-k – gives degree of freedom (d.f)

This value will be obtained from any t- table of statistical books. From the table can be found
2
in the top of row & the d.f. can be obtained in the first columns of that table. Then the point

where (significance level) & the d.f. are interesting with each other gives you the table value
2
of the t- test.

C) The last stage will be compare a & b i.e. the calculated value with the table value.

Now one should be aware of the following idea i.e the computed value of t should be calculated
for each estimator but you will have only one table value of t that will be obtained from the
statistical table. Then each calculated value of t from the estimators should compare with the
fixed table value of t & the procedure of comparison is.
1) If the calculated value of t test is less than the table value of t then accept the null hypothesis
& reject the alternative. It means
 ˆ  
If t   < t , (N − k)
 S .E.ˆ  2

Haramaya University 59 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 ˆ  ˆ
  t  , (N − k)
t 
ˆ 
 S .E (  )  2
We accept the null hypothesis that Ho= ˆ = 0
Ho= ˆi = 0
Reject the alternatives that H1 = ˆ  0 & H 1 = ˆi  0

Means
i. There is no relationship between Y& Xi's or the dependent variable Yi is not explained or
depend on the in depend X1 & X2
ii. ˆ & ˆ1 , ˆ2 are not different from zero or they are in significant

2) If the calculated t- value is greater than the table t- value we reject the null hypothesis
i.e. Ho= ˆ = 0 & Ho= ˆ = 0 and accept the alternatives i.e. H1= ˆ  0 & H1= ˆ  0
 ˆ  
If t    t , (k − N )
 S .E (ˆ )  2
 ˆ 
t    t  (k − N )
ˆ 
 S .E (  )  2
In this case we can interpreter it as follows:
i. the estimates are significantly different from zero i.e  ˆ & ˆ1 ˆ 2 are not equal to zero
ii. the estimates (  ˆ & ˆ ˆ ) are statistically significant
1 2
iii. the explanatory variables X1 & X2 will affect the dependent variable Yi.

Unadjusted R2, (Adjusted) Corrected R 2 & F-test

R 2 is the coefficient of determination or called the unadjusted R 2 . This unadjusted coefficient of
determination will have the same meaning just like in the two variable cases the only difference
here is that the inclusion of additional variables. Then R 2 explain the proportion of the variation
in Y explained by the variables X1 & X2 jointly.
R 2 Can be derived from equation no 3.1 just by following the same procedures of single
equation model calculation of r2.
ˆ ˆ
ESS  1  yi x1 +  2  x2 y
R2 = =
TSS  yi 2
If you compare this equation with the previous r2.
 xiyi
ˆ
2
r =
 yi 2 the only difference is we added ˆ  x
2
2
y because we incorporate the

3rd variable X2i. Then as you increase your explanatory variables the value of R2 will increase
because as there numerator is increasing the denominator remains the same. Due to this reason
we call it unadjusted R2. Therefore, if we have two models with different explanatory variables,

Haramaya University 60 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

we cannot compare them on the basis of R2. But to compare two equation with different number
of explanatory variables we should have to use Adjusted (corrected) R 2 .

Adjusted R 2
 iuˆ n−k
2

R 2 =1 -
 yi 2
n −1
Where n-sample size
k-degree of freedom

OR
n −1
R 2 =1-(1- R 2 )
n−k
If the sample size is small R < R 2 but if the sample size is very large R 2 & R 2 will be very close
2

to each other’s. Here we should be aware of that for very small size of sample R 2 may be
negative but taken as zero. Note that if R 2 =1, R 2 =1, when R 2 =0, R 2 = (1-k)/(n-k) in this case
R 2 will be negative if k>1.

F- tests:-
Under this task we compare the computed F-values with the table value of F-tests. Importance of
–F- tests: - This test is undertaken in multiple regression analysis for the following reasons.
a) To test the overall significance of the explanatory variables i.e. whether the explanatory
variables X1,X2 actually do have influence on the explained variable Yi or not.
b) Test of improvement, means by introducing one additional variable in the model to test
that whether the additional variable will improve the influence of the explanatory variable
on the explained variable or not.
Ex. Yt =  + 1 X 1i +  2 X 2i + Ui
Yt =  + 1 X 1i +  2 X 2i + X 3i + Ui
Now if you take the first equation it has only two explanatory variables X1 &X2 but in the second
we included X3. Then the addition of X3 may affect positively or negatively the relationship
between Y & X1, X2.
c) To test the equality of coefficients obtained from different sample size (chaw test).
Ex. Suppose you may have a sample data of agricultural output of Haramaya woreda
from 1974 E.C. up to 2010 E.C. Now if you want to know the change in agricultural
output before & after the fall down of Derge (1983 E.C). By splitting these data in to two
you can compare the coefficients & by doing so you can undertake F-test and see whether
there is a change in agricultural output or not.
d) Testing the stability of the coefficients of the variables as the number of sample size
increase
Ex. You may take first a 10 year sample for your study & estimate your estimators. But if
you increase the number of sample size in to 15 years, will the coefficients of the
variables are stable or not will be tested using F-tests.

Haramaya University 61 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

F test can be calculated for a given estimated equation using the following formula.

F* =
 yˆi / k − I = R 2 /(k − 1)
 ei 2 / N − k (1 − R 2 ) /( N − K )
Table value of F-test can be read from the statistical table as follow.
Fv1 , v 2 or F( k −1), ( N −k )
Where V1= (K-1) & V2= (N-K) both of them explain degrees of freedom V1 the numerator & V2
the denominator degree of freedom. To undertake F test we should compare the calculated value
with the table value.
1) If the calculated F value is less than the table value i.e.
R 2 / k − 1)
*
F =  FV 1,V 2
(1 − R 2 / N − k
Where F* is calculated F value
F is the table value
It means we accept the null hypothesis that H0= ˆ = 0 & H0= ˆi = 0 and reject the alternative
hypothesis which says H0= ˆ  0 & H1= ˆi  0 . The interpretation will be
a) The estimators ( ˆ , ˆ , ˆ2 ) are equal to zero then the estimates are insignificant.
b) The explanatory variables (X1,X2) do not have influence on the explained variable Yi.
2) If the calculated F value is greater than its table value i.e.
R 2 / k − 1)
F* =  FV 1,V 2
(1 − R 2 / N − k
It means we reject the null hypothesis that H0= ˆ = 0 & H0= ˆi = 0 & accept the alternatives that
H1= ˆ  0 & H1= ˆi  0 . The interpretation of this is just the opposite of a & b in the above
sentences.
Example:- Suppose the quantity supplied of commodity is assumed to be a linear function of the
price of the commodity itself & the wage rate of labor used in the production of the commodity-
If the supply equation is given by
Q1 =  + 1 PX 1 +  2W + u
Where Q1 = quantity supplied. Px, price of commodity X & W is wage rate.

Example: Using the following sample data

a) Estimate the parameters using OLS
b) Test the statistical significance of the individual coefficients( ˆ , ˆ & ˆ2 ) at 5% significant
level
c) Test the overall significance of the coefficients (F-test)
d) Compute the price elasticity of supply

Y Q1 20 35 30 47 60 68 76 90 100 105 130 140 125 120 135

X Px 10 15 21 26 40 37 42 33 30 38 60 65 50 35 42
Z W 12 10 9 8 5 7 4 5 7 5 3 4 3 1 2

Haramaya University 62 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

From these we will have the following values.

Let Qs =Y, Px = X & W=Z
Y = 85.4 x =36.27  yx =7207.4  Zy =-1553  xz =-514.667  yi 2
=23211.6
 xi 2
=3192.93 Z 2
=135.3  ei 2
4386.49 N=15

Answer
a) Estimate the parameters using OLS

ˆ1 =
 x yi z −  xz zy
1
2

 xi  z − ( xz )
2 2 2

(7207.4)(135 − 3) − (−514 − 667)(−1553) 173668

= = 1.16
(3192.93)(135.33) − (−514.667) 2 162,712

ˆ2 =  zy x −  Xz xy

2

 xi  Z − ( xz )
2 2 2

(−1553)(3192) − (−514)(7207)
=
162,712
(−495,7176) − (−3704398) − 1,252778
= =  −7.69
162712 162,712
ˆ = Y − ˆ1 x − ˆ2 Z
85.4-(1.06x36.2)-(-7.699x5.6)=
̂ =85.4-38.372-(-42.946)=89.974
Y = ˆ + ˆ1 x − ˆ2 Z
Y= 89.974 + 1.06X - 7.69Z.
This equation will be read as follow:
✓ ̂ = 89.974 means if the price of the commodity & the wage rate is zero the supplier will
supply 89.974 units of goods. But it is meaningless to interpret the constant term ( ̂ ). (In
some analysis it doesn’t give sense.)
✓ ˆ1 is the coefficient of price of the commodity. The value 1.06 signifies that if the price
of the commodity is increasing by one birr given the price of wage is constant quantity
supplied will increased on the average by 1.06 units.
✓ ̂ 2 is the coefficient of the wage rate. If the wage rate is increased by 1 birr keeping
constant the price of the commodity quantity supplied will decrease on the average by
7.69.
✓ ˆ1 & ̂ 2 are coefficients of the explanatory variables & they are containing the marginal
values. If you take the first derivative of the equation, you will have marginal values.

Ex. Ŷi = 89.974+16px-7.697Z

yi 2Yˆi
= 1.16 & = −7.697
px 2Z

Haramaya University 63 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

b) Test the statistical significance of the individual coefficients ( at 5% significant level

Standard error test
To test this we have to have variance & the standard error of the parameters ( ˆ , ˆ & ˆ ) .The 2

variance of ˆ , ˆ & ˆ2 will have a value of  2 (see equation number 4.16 – 4.18).

Where u 2 =
 ei 2

from our example n=sample size of 15

n−k
K is number of estimated parameter which is 3( ˆ , ˆ & ˆ2 )

u 2
=
 ei 2
4386.49
=
= 365.5
N −k 15.3
Using equation number 4.16 we can calculate Var (ˆ )
1  (36.2) 2 (135) + (5.6) 2 (3192) − ()2 x36.2 x5.6 x − 514) 
Var (ˆ ) = +  365.5
5  162712 
 (0.0666 + 176,909.4 + 100101 + 207244.8 
= 365.5
 162712 
 (0.0666 + 484,255.2 
Var (ˆ ) =  365.5 = 1,110.54
 162712 
ˆ
Var ( 1 ) again we can calculate using equation number 4.18
 135.33 
Var ( ˆ1 ) = 365.5   = 0.303
162,712 
Var ( ˆ ) = the Var of ˆ can be calculated using equation number 4.19
1 1

 3192.933 
Var ( ˆ2 ) = 365.5   = 7.17
 162,712 
From the above values we can calculate S.E (ˆ ) , S.E ( ˆ1 ) & S.E ( ˆ2 ) as follows
S.E (ˆ ) = Var̂ = 1110.54  31.748
S.E ( ˆ1 ) = Var ( ˆ1 ) = 0.303  0.521
S.E ( ˆ2 ) = Var( ˆ2 ) = 7.17  2.529
Having calculated S.E of the coefficients of the variables ( ˆ , ˆ & ˆ2 ) we can undertake S.E.
tests – as follows
̂
If S.E( ̂ ) > we can accept the null hypothesis & reject the alternative
2
ˆ 89.974
S .E (ˆ ) = 31.76 & = = 44.987
2 2
̂
Then S.E (ˆ ) =i.e 31.76 is less than , which is 44.987. Therefore, we can conclude that ̂ is
2
statistically significant

Haramaya University 64 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

ˆ 1.16
S.E ( ˆ1 ) = 0.521 1 =0.58
2 2
ˆ
From this we can see that S.E ( ˆ1 ) is less than 2 then we can conclude that ( ˆ2 ) is significant.
2
ˆ 7.697 ˆ
S.E ( ˆ2 ) = 0.529 & 2 = = 3.848, Again here S.E ( ˆ2 ) is less than 2 .
2 2 2
All the estimators are statistically significant or we reject the null hypothesis that H0= ˆ = 0 ,
H0= ˆi = 0 (we reject the hypothesis which says ( ˆ , ˆ & ˆ2 ) are equal to zero) & accept the
alternative that H1= ˆ  0 , ˆ  0 H1= ˆi  0 ( ˆ , ˆ & ˆ ) are different from zero.)
i 2

The economic interpretation of rejecting the null hypothesis and accepting the alternative states
the following:
i. The estimators are statistically significant; and
ii. The explanatory variables X1 & Z (price of commodity X & wage rate) influence the supply
of commodity (Y).

Student –t –test
In the t-test analysis we compare the calculated t with table value of t. How to get calculated t-
value
ˆ 89.974
t= Computed value of t = =2.23
S .E (ˆ1 ) 31.76
ˆ1 1.16
t= Computed t = =2.83
S .E ( ˆ ) 1
0.521
ˆ − 7.697
t= 2
Computed t = =3.043
S .E ( ˆ2 ) 2.529

How to get t-value from t- table

Given 5% significance level, the total sample used to calculate the estimators are 15 & the
number of estimators is 3 (ˆ , ˆ & ˆ2 ) . If we take two tail test

t ,(N − K )  = Significance level
2

Show two tail test
2
N = sample size
K = estimators
N-K = degree of freedom
0.05
t , (15 − 13) = t0.25,12
2

Haramaya University 65 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

From the t- table in the top of the raw find 0.025 & in the first column of the table find 12 then
when these two values are intersecting with each other, that point will give you the table value of
t. From our t 0.025,12 the table value is 2.179. Compare this table value with the computed value
& if the computed value is greater than the table value we reject the null hypothesis & accept the
alternative. Again if the computed value is less than the table value we accept the null hypothesis
& reject the null hypothesis.
Compare computed with table value
Compared t value of ̂ is 2.83 & the table value is 2.179 here the computed t value is greater
than the table value for ̂
Again the computed t value for ˆ & ˆ is 2.23 & 3.043 respectively they are greater than the
1 2

table value.
Since in all this cases ˆ , ˆ & ˆ2 computed t-value is greater than the t-table value. We will have
the following interpretation
i. ˆ , ˆ & ˆ are statistically significant
2

ii. The quantity supplied is influenced by the price of the commodity & wage rate

Coefficient of determination /R2/

R2 =
ˆ 1
 yi x1 + ˆ  yzi
(1.16)(7207.4) + (−7.47)(−1553)
2
=
 yi 2
23211.6
(8,360.58 + 11,600.91) 19961.49
R2 = = = 0.8599
23,211.6 23,211.6
This means 85.99% of quantity supplied is explained by price of the commodity & wage rate.

Adjusted R 2
n −1 15 − 1
R 2 =1-(1- R 2 ) = 1-(1-0.8599) = 0.8365
n−k 15 − 3

F – test
The overall significance of the explanatory variables can be tested using F-test. Just like t- test in
the case of F test we will have computed & table value of F. Calculated value of F* can be
obtained using the following formula
R2 / k − 1 0.8599 / 3 − 1 0.8599 / 2 0.42995
F* = = = = =36.826
(1 − R ) / N − k 0.1401/ 15 − 3 0.1401/ 12 0.11675
2

Table value of F can be obtained by taking the enumerator degree of freedom (k-1) & the
denominator degree of freedom (N-K). Then F (K-1), (N-K). Given the level of significance of
5% we can get F2, 12 i.e. K-1 = 3-1 = 2 & N-K = 15-3 = 12. Then from the table we find
the enumerator degree of freedom (K-1) in the top row of the table & the denominator degree of
freedom (N-K) in the first column of the table at the intersection point of these values you will

Haramaya University 66 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

get the table value. From our example F2 , 12 at 5% level is 3.89. Comparison of calculated F &
table value of F. If the calculated value of F is greater than the table value then we can reject the
null hypothesis & accept the alternative. From our example the calculated F value is 36.826 is
greater than the table value 3.89. The economic interpretation of this is:
• All the estimators are significant or statistically different from zero
• Quantity supplied is affected by the price of the commodity & wage rate
Presentation of regression results:- Different books used different presentation methods but the
most commonly is the one which we write under here using our previous example.

Y = 89.974 + 1.16PX - 7.697Z

S.E = (31.748) (0.521) (2.529)
t= (2.83) (2.23) (3.043)
R2 = 0.8599
R 2 = 0.8365
F = 36.826
N = 15
 ei 2 = 4019.714
4.4. Importance of the statistical Test of Significance
There is no general agreement among econometricians as to which of the two statistical criteria
is more important a high R2 ( R 2 ) a lower standard error of the parameter estimates
The choice would not be difficult if the model produces high R2 and lower S.E. However, this is
not usually the case in most applications we found high R2 and higher or insignificant standard
errors.
o In this event some econometrician tends to attach great importance for R2 and accept the
parameter estimates; despite the fact that some of them are statistically insignificant.
o Other suggests that acceptance or rejection of the estimates which are not statistically
significant depends on the aim of the model.
i. If the aim of the model is forecasting majority of econometricians attach importance for high
R2.
ii. If the aim of the model is for policy analysis of economic phenomena great importance is
attached for lower S.E.
A high r2 has clear merit only when it is combined with significant estimates (low standard
errors). When high R2 & low S.E are not found in any particular study the researcher should be
very careful in his interpretation & acceptance of the results. Priority should always be given to
the fulfillment of the economic a priori criteria (sign & magnitude of the estimates). Only when
economic criterions are satisfied should one proceed with the application of different tests.

Haramaya University 67 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

4.5. Exercise for chapter four

1) From the following data compute the regression of automobile expenditure on consumer
expenditure & other travel expenditure (Take the linear regression analysis)

Automobile
212 158 180 253 175 429 437 419 318 355
expenditure
Other travel
29 46 28 26 29 64 119 81 74 66
expense
Consumer
2437 2476 2132 2256 2258 3566 4486 3602 3446 3736
expenditure

a) Calculate the coefficient of parameters using OLS regression equation of consumer

expenditure on other travel expense & other travel expense?
b) Write the equation
c) Test the significance of the parameters using standard error & t-test?
d) Construct the 95% confidence interval for the parameters?
e) Calculate the unadjusted & adjusted R2. Why the difference is arising?
f) Test the overall significance of the regression?
g) If the relationship is nonlinear how would you interoperate the results (coefficients)
h) Calculate the partial correlation coefficients of the parameters & interoperate the results?

2) Given the following data & answer the question from question 1 (i.e. from a up to h)

Year 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Y 40 45 50 55 60 70 65 65 75 75 80 100 90 95 85
X1 9 8 9 8 7 6 6 8 5 5 5 3 4 3 4
X2 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800

Y= quantity demanded,
X1 = is the price of the commodity &
X2 is consumers income
Assume that income is the dependent variable & the remaining are the independent variable.

Haramaya University 68 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Chapter 5: Relaxing the assumptions of the classical model

5.1. Introduction
In the previous chapters to estimate the coefficients of variables in the two variable and more
than two variable cases we utilize two basic assumptions, one about the random term Ui & the
other about X (independent variable). These are:

o The first assumption about the random term Ui

a) Ui is a random term. We introduce this random term to capture those variables which are not
incorporated (omitted) in the model, misspecification of the model, error committed in the
measurement of variables & the erratic behavior of human beings.
Though we incorporate Ui to solve the above problems, there is no formal test for this
random term measurement. Ui's are not observable & their estimates e's are obtained with the
assumption of randomness.
b) Assumption of zero mean of Ui. We have said that the expected value (average value) of Ui
is equal to zero E(Ui)=0 In reality it is impossible to prove or disprove it; but it is necessary
to make the zero mean assumption so as to be able to apply the rule of algebra to stochastic
phenomena & nature of relationships. Otherwise it would be impossible to estimate with the
common rule of mathematics.
If E (Ui)> 0 or E (Ui) < 0 then the estimated line would be biased. This is the basic reason why
the assumption of E (Ui) =0 is forced up on us if we are establish the true relationship.
c) The assumption of homoscedasticity:- we assume that the distribution of the random term Ui
is remains the same over all observation of the explanatory variables; in particular the
variance is constant. Var (U)=  2 constant. This assumption is known as homoscedasticity.
But if this assumption does not holds true what will be the var (Ui) or if the var (Ui)   2ui
i.e. if the variance of Ui is not constant. This is called hetroscedasticity
d) Assumption of no- autocorrelation or No- serially correlated. We assume that cov (UiUj)=0
i  j. i.e. the covariance between the successive random term Ui is zero or the successive
random terms are not influenced with each other. i.e. Ut (current random term) occurring will
not be affected by Ut-1 (previous year random term). But there is a chance the Ui's to be
influenced with each other’s i.e. if Ut is influenced by Ut-1 cov(UiUj) = 0 (There is
autocorrelation or Uis are serially correlated.

o The 2nd Assumption about X’s (explanatory variables).

In the estimation of the model using OLS method we assume that the explanatory variables are
not perfectly correlated with each other’s & we call this there is no- multi co linearity. But in
reality the X’s may be correlated with each other’s & problem of multicollinearity may be
observed in the model. Now the question is that if we violate these two assumptions about Ui &
Xi’s what will be the consequence on the reliability of our estimated numerical values of the
coefficients of variable (estimates ) or will the estimates still contains the desirable properties of

Haramaya University 69 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

OLS estimates? The answer is short and precise that is the estimates will not satisfies the
desirable properties of OLS.

5.2. Violation of the important assumptions

5.2.1. Hetroscedaticty:-
If the probability distribution of Ui remains the same over all explanatory variables this
assumptions is called homosceasticity i.e. var (Ui)= u 2 constant Variance. In this case the
variation of ui around the explanatory variables is remains constant. But if the distribution of ui
around the explanatory is not constant we say that ui’s are hetro scedastic (not constant
variance).Var (ui) = u 2 i. signifies the fact that the individual variance may be different.
The assumption of homoscedasticity states that the variation of each random term (Ui) around its
zero mean is constant and does not change as the explanatory variables change whether the
sample size is increasing, decreasing or remains the same it will not affect the variance of Ui
which is constant.
Var (Ui) = u 2  f(Xi)-------------------------------------------------------5.1
This explains that the variation of the random term around its mean does not depend upon the
explanatory variable Xi. This constant variance is called homoscedastic (constant Variance)
But the dispersion of the random term around the regression line may not be constant or the
variance of the random term Ui may be a function of the explanatory variables. Var (Ui) =
 2ui = f ( Xi) here i- signifies the individual variance may all be different. This is called
hetroscedasticity or not constant variance. The case of hetroscedasticity is shown by the
increasing or decreasing depression of the random term around the regression line as shown in
fig b, c & d.

Y Y

X X
Fig. (a) Fig. (b)

Y Y

X
X

Haramaya University 70 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Fig. (c) Fig. (d)

On diagram (a) you can see that the random term Ui is dispersed with in a constant variance
around the regression line.
In figure (b) as the value of X is increasing the variance of the random term Ui is also increasing.
Ex1.
St =  + Yt + Ut ---------------------------------------------------------5.2
Where St is saving, Yt is income  &  parameter & Ut is random term.
If you wants to estimate this saving function & collect cross sectional data from a lower income
group level the variation of saving is lower where as there is greater variation in the saving
behaviors of high income-families. Thus the random term Ui is very low at lower income and it
tends to increase as income increases due to variation in the behavior of saving.
Ex2. Suppose we try to study the consumption expenditure from a given cross sectional sample
of family budget.
Ct =  + Yti + U ----------------------------------------5.3
Where Ct= consumption expenditure
Yt= disposable income of the h.h
Again at a lower income level the consumption expenditure is almost equal i.e. no variation in
consumption expenditure but at a higher level of income there is variation in consumption
between higher & lower income level. Then there is a possibility of increasing variation of the
random term called hetro scedasticity. In figure (c) we can see that as the value of X is increasing
the variation of random term Ui is decreasing. By doing & learning the error of committing
errors will decrease. In this case the variation of the term Ui is decreasing as Xi's is increasing.
Again here we will have hetroscedasticity. In figure (d) we may have a complicated
hetroscedasticity. i.e. in the beginning there is a high variation of U at lower Xi's & a higher
variation of Ui at a higher Xi's.

5.2.1.1. Consequences of hetroscedasticity

If the assumption of homoscedasticity is violated, it will have the following consequences
1) The variance of the coefficients of OLS will be incorrect & under the assumption of
homoscedasticity we have
2 
X2
ˆ
var ( ) = u
2

 xi 2
1
Var ( ( ˆi ) = u 2
 xi2
Since the variance of  2 is assumed to be constant we took it out from the summation in
equation number 2.47 and 2.48 But under hetroscedasticity condition  2 is not constant & we
will have
2 
x 2u 2i
ˆ
var ( ) = u &
 xi2
Var ( ( ˆi) = ki 2ui


Haramaya University 71 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 2ui in this case is not a constant number but varies as X changes. To calculate var (ˆ ) and
var ( ˆi ) we should know the value of  2ui i.e. i=1, 2...n. The problem encountered here is that
since we can not have observable value of Ui we have to estimate it from the sample data using
 
residuals as proxies (Yi − Yˆ ) = ei to the unobservable errors. Then we have to estimate n
variables from n- observables i.e. one for each variance, a situation in which estimation is
impossible.
OLS estimators shall be inefficient: - If the random term Ui is hetroscedastic, the OLS estimates
do not have the minimum variance in the class of unbiased estimators. Therefore they are not
efficient both in small & large samples.
In case of hetroscedasticity
var(ˆ ) = ui
2

 xi2
=  ki 2u 2
=  ki 2 E(Ui) 2
2
 xi 
=   E (ui) 2
  xi 2 
 
ˆ
var( ) =
 xi 2
. 2ui Under hetroscedasticity.
( xi )2 2

The variance of ( ˆ ) under homoscedasticity is

ˆ  2u
Var (  ) = Since  2u is constant =  2u 1
 xi 2
 xi 2
Now Var ( ˆ ) under hetroscedasticity the variance of Ui is proportional to Ki (increasing,
decreasing or a combination of both. i.e. what ever its relation with the X is given by this
proportion of Ki) then we will have
 2ui = ki 2 ------------------------------------5.4

This means the hetroscedasticity is a proportion of (Ki) the homoscedasticity & Ki is non-
stochastic constant weight.
Substitute  2ui = ki 2 in the variance of ( ˆ ) when it is hetroscedastic

ˆ ) =  xi ui  xi (ki
2 2 2 2
u)
(  =
( xi ) ( xi )
Var 2 2 2 2

=  2u
 xi ki
2

( xi )
2 2

Var ( ˆ ) =
 xi ki 2

=
 u 2   xi 2 ki

  xi 2  ( xi 2 )
 xi . xi
2 2
 
From the equation we will have two components.

Haramaya University 72 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 is variance of ( ˆ ) under homoscedastic assumption & 

 u 2  kixi2

  xi 2 
   xi 2
If x & ki are positively correlated then
 kixi2
is greater than 1.
 xi 2
The var ( ˆ ) under hetroscedasticity will be greater than its variance under homoscedastic.
Following this true standard error of ˆ shall be underestimated & the t- value associated with it
will be overestimated.
This leads to the conclusion that in a specific case at hand ˆ is statistically significant (which in
fact may not be true).
The consequence of var ˆ of hetroscedastic is that the confidence limits & the test of
significance will not be applicable
If we proceed with our model under false of homogeneity variances then our inference &
predictions about the population coefficients would be incorrect.

5.2.1.2. How to detect Hetroscedasticity

Hetroscedasticity have a serious effect on OLS estimates. If so how does one know the existence
of hetroscedasticity in the model? Various tests have been suggested for establishing
hetroscedasticity & some of them are described below.

Test 1. The spearman rank- correlation Test

This is the simplest & approximate test for defecting hetroscedastic which will be applied either
to small or large samples & may be outlined as follows.
we regress Y on X
Y =  + Xi + Ui
Obtain the value of residuals ei = Yi − Yˆ which are estimate of the U’s. Then arrange ei’s (ignore
the sign or take the absolute value of ei) in ascending or descending values of X & we compute
the rank correlation coefficient.
6 D 2
r e− x = 1 −
1
− − − − − − − − − − − − − − − − − 5.5
n(n 2 − 1)
Where: Di is the difference between the ranks of pairs of corresponding pairs of X & ei & n is
the number of sample observations.
A high rank correlation coefficient suggests the presence of hetroscedasticity. If we have more
explanatory variable we may compute the rank correlation coefficient between ei & each one of
the explanatory variables separately.

Test 2. The Goldfeld Quandt test

This test is applicable for large samples & the number of observations (at least) i.e sample size is
twice the number of explanatory variables
Yi =  + Xi +  2 X 2i +  3 X 3i + Ui

Haramaya University 73 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

The numbers of explanatory variables are 3(X1, X2, X3) then the sample size is at least must be
6.In addition to the size of sample this test assumes normality & no autocorrelation. The steps
described are as follows
Formulate hypothesis testing.
The null hypothesis H0 =Uo are homoscedastic
The alternative H1= U1 are not homoscedastic (U’s are hetroscedastic.)
Order the observations according to the magnitude of the explanatory variable X.i
Omit certain number of central observations say C amount. Now if you deduct the amount that
will be omitted C we left with n-c number of observations. Then divided this remaining sample
n−c
in to two parts. i.e . These samples contain two pars one part includes the small values of
2
X while the other parts the large values of X.
Fit separate regression by OLS procedure & obtain the sum of squared residuals from each of
them
Let  e1 is the residual squared from the sample of low values of X &  e2 2 from the large
2

sample values of X. Then calculate F- test using ratio of the residuals variances.
e
2
2

 n − c  
  − k
F =
*  2   = e 2
2
− − − − − − − − − − − − − − − − − 5.6
e 2
1 e 2
1
 n − c  
  − k
 2  
n−c
The numerator degree of freedom is   -k & the denominator degree of freedom is
 2 
n−c
  -k. From this one can see that the numerator is equal to the denominator i.e. V1=V2.
 2 
Using V1&V2 we can find the table value of F v1,v2 & compare this table value with the above
calculated value.

If
 e22
 Fv1,v2 if the calculated value is greater than the table value of F then our decision
 e12
will be
Reject the null hypothesis that Ho=Ui are homoscedastic
Accept the alternative that H1=Ui’s are hetroscedastic. Then the variance of Ui’s are not constant
(homoscedastic) but the variance of Ui’s are hetroscedastic.

If
 e22
 Fv1,v2 the above decision will be reversed.
 e12
Ex. Given the following hypothetical data on consumption expenditure Y & income X.
Rank data in ascending order of X

Haramaya University 74 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Y X Y X
55 80 55 80
65 100 70 85
70 85 75 90
80 110 65 100
79 120 74 105
84 115 80 110 Lower values of X’s
98 130 84 115
95 140 79 120
90 125 90 125
75 90 98 130
74 105 95 140
110 160 108 145
113 150 113 150
125 165 110 160
108 145 125 165 Middle observations C=4
115 180 115 180 that will be omitted
140 225 130 185
120 240 135 190
145 185 120 200
130 220 140 205
152 210 144 210
144 245 152 220
175 260 140 225 higher values of X’s
180 190 137 230
135 205 145 240
140 265 175 245
178 270 189 250
191 230 180 260
137 250 178 265
189 275 191 270

Take the first 13 observations i.e lower observations & run regression

Y =  + Xi + Ui
Then you will find
Y =3.4094 + 0.6968Xi
(8.7049) (0.0744)
r2= 0.8887
 e1 = 377.11
2

Haramaya University 75 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Again from the remaining observations 13 sample run regression Y on X & you will have

Yˆ − 28.0272 + 0.7941xi
(0.1319)
r2= 0.7681
 e2 = 1536.8
2

Computed value of F =
e 2
2
=
1536.8
= 4.07
e 2
1 377.11
Table of Fv1,v2
n−c  30 − 4 
v1 =  −k =   − 2 = 11
 2   2 
n−c  30 − 4 
v2 =  −k =   − 2 = 11
 2   2 

F11,11, from the table of F at 5% level & you will get 2.82. Compare the F table value with F
computed value. If the computed value is greater than the table value rejects the null hypothesis
that there is homoscedasticity & accepts the alternative that there is hetroscedasticity. The
computed value 4.07 is greater than the table value i.e 2.82. Then we can say that there is
hetroscedasticity.

Test 3 the Glejser- test for homoscedasticity

This test can be outlined as follows
We perform the regression of Y on the explanatory variables ( X) & we compute the residuals
i.e. ei.
Yi =  + X 1i +  2 X 2 i +  3 X 3i + Ui − − − − − − − − − − − − − 5.7
Estimate this you will have
Yi = ˆ + ˆX 1i + ˆ2 X 2 i + ˆ3 X 3i + Ui − − − − − − − − − − − − − 5.8
Find
ei = Yi − Yˆi − − − − − − − − − − − − − − − − − − − − − 5.9
Regress /ei/ on each explanatory variables Xi’s independently using different power.

/ e / =  0 + 1 Xi 2 + vi
/ e / =  0 + 1 Xi + vi
1
/ e / =  0 + 1 + vi -------------------------5.10
X
1
/ e / =  0 + 1 + vi e.t.c.
X

Haramaya University 76 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

But here it is assumed that vi satisfies all the assumptions of OLS. We choose the form of
regression which gives the best fit in light of
R2 & the standard error of coefficients of ˆ & ˆ
Formulate the hypothesis testing
The null hypothesis H0 = U0 s are homoscedastic
The alternative hypothesis is H1 = Ui’s are hetroscedastic.
Using standard error test we can accept or reject the null hypothesis as follows
ˆ ˆ
If S .E (ˆ )  & S .E ( ˆ )  we reject the null hypothesis (the existence of homoscedasticity is
2 2
rejected) & accept the alternatives which says that there is hetroscedasticity. This kind of
hetroscedasticity i.e if we reject the null hypothesis of ˆ & ˆ & accept the alternative is called
mixed hetroscedasticity.
ˆ ˆ
If S.E ( (ˆ )  & S .E ( ˆ )  i.e we accept the null hypothesis for ̂ & reject the null
2 2
hypothesis for ̂ this is called pure hetroscedasticity.
In addition to the standard error test we use the F & t-tests of the significance for the coefficients.
On the basis of t & F test if the ̂ & ˆ are significantly different from zero we reject the null
hypothesis & accept the alternative and concluded that there is hetroscedasticity.
One of the advantages of the Glejser test is that it gives the form of hetroscedasticity i.e. what is
the relationship between  2 ui & f ( X ) i.e increasing or decreasing relationship etc. This test is
very important in correcting or removing hetroscedasticity.

5.2.1.3. Solutions for Hetroscedasticity

When hetroscedasticity is found in the model under consideration the appropriate solution is
Transform the original model so as to obtain a homoscedastic disturbance term i.e constant
variance. Then we apply the method of CLS to the transformed model.
The transformation of the model will change the original data of values of explanatory variable.
The transformation of the model depends up on the nature of hetroscedasticity that will be
explained by the functional relation ship between the variance of the disturbance term ui 2 and
the values of the explanatory variables (X’s)

 2 ui = f ( X ) − − − − − − − − − − − − − − − − − − − −5.11
The transformation of the original model consists in dividing through the original relationship by
the square root of the term which is responsible for hetroscedasticity. It means simply dividing
the original model by the square root of the variable which is identified or responsible for the
existence of hetroscedasticity. Suppose lets have the following type of original model.

Yi =  + Xi + Ui − − − − − − − − − − − 5.12
Where Ui satisfies all the assumptions but Ui is hetroscedastic
Ui  N (0,  2 ui) − − − − − − − − − − − −5.13

Haramaya University 77 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

E(Ui) 2 =  2ui = f ( Xi) . When there is hetroscedasticity there is functional relationship

between  2 ui = f ( Xi) . But the problem is what the exact functional relationship between these
two. Practically it is impossible to obtain the actual form of relationship but we can only assume
possible types of hetroscedastic structures.
Case (a) suppose the hetroscedastic type is the form of

E(Ui)2 =  2 ui = Ki 2 Xi 2 ------------------------------------5.14
Where Ki 2 is a finite constant term to be estimated from the model. It explains the variance of
the random term increasing by Ki 2 (proportionately) as the explanatory variables increases by
Xi 2
 2ui = Ki 2 Xi 2
Solve for Ki 2
 2 ui
Ki 2 = − − − − − − − − − − − − − − − − − − − −5.15
Xi 2
This suggests that the appropriate transformation of the original model is the division of the
original relationship by X 2
The reason for this is that
ui 2 ui 2
K =
2
= Ki =
Xi 2 Xi 2
ui
K = ------------------------------------5.16
Xi
This shows divided the orginal model by Xi or X 2 then
Yi  Ui
= + − − − − − − − − − 5.17
Xi 2 Xi 2 Xi 2 i
Yi  Xi Ui
= + + − − − − − − − − − − − − − −5.18
Xi Xi Xi Xi
Ui
From this equation the new transformed random term is homoscedastic (constant variance).
Xi
Ui
To prove this is homoscedastic.
Xi
2
 Ui   Ui  1
Var  = E   = 2 E (Ui) 2 − − − − − − − − − − − − − 5.19
 Xi   Xi  Xi
We know that E(ui) =  2ui substitute in equation number 5.13
2

 Ui  1
Var   = ui 2 ----------------------------------------5.20
 Xi  Xi
2

The type of hetroscedasticity we assume was that just like in equation 5.15 then substitute it in
equation number 5.18 & you will get

Haramaya University 78 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 Ui  1
Var  = 2 (k 2 X 2 )
 Xi  Xi
 Ui 
Var  = ki 2 is a constant number which proves that the new random term in the model have a
 Xi 
finite constant variance Ki2. Then we can apply OLS to the transformed original model of
equation number (4.16)

Yi  Xi Ui
= + +
Xi Xi Xi Xi
Yi 1 Ui
= + + − − − − − − − − − 5.21
Xi Xi Xi
In this transformed model the position of the coefficient has changed i.e. the constant term  in
the original model equation number 5.12 now will be the coefficient of the 1 Xi & the  which
was the coefficient of Xi in the original model appear to be a constant term in the transformed
model of equation (5.21). Then if you want to get the original model multiply by Xi the
transformed model (equation 5.21).
Case (b) Suppose the form of hetroscedasticity is

E (ui)2 =  2ui =k2Xi -------------------------------4.22

It means as X is increasing or decreasing the variance of the random term Ui is increasing or
decreasing by K2. Then from equation number 4.22 we can get the value of K2
ui 2
K2 =
Xi
ui
K = − − − − − − − − − − − − − − − − − −4.23
Xi
This means the original model equation number 4.11 will be divided by Xi

Yi  Xi Ui
= + = − − − − − − − − − 4.24
Xi Xi Xi Xi
or
Yi  Ui
= +  Xi + − − − − − − − − − 4.25
Xi Xi Xi
Ui
The disturbance term in the transformed model is homoscedastic
Xi
2
 (Ui 
Var (Ui ) = E 
1 1
 = E (Ui) 2 = ui 2 − − − − − − − 4.26
 Xi Xi
 Xi 
We have assumed that in equation number 5.22 the type of hetroscedasticity is substituted in
equation number 5.25. Then

Haramaya University 79 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Var (Ui ) =
1 1
ui 2 = ( K 2 X ) = K 2 − − − − − − − − − − − − − 4.2.7
Xi Xi
Var (Ui) = K2 shows that the variance of the random term Ui after transformation of the original
data will give us a constant number equal to K2. Therefore we can apply OLS to equation
number 4.25. In the transformed model we do not have intercept term then the equation pass
through the origin to estimate ˆ & ˆ . In this model if you wants to get the original model we
shall have multiply the transformed model by Xi .
Case (C) suppose the form of hetroscedasticity is the form of
E(Ui 2 ) = ui 2 = K 2 ( E(Yi)) 2 − − − − − − − 4.2 8
In this equation we assumed that the variance of the disturbance terms is proportional to the
square of the dependent variable Y.


Var (Ui2)=  2 ui = K 2 (E (Yi) ) = K 2 ( + Xi) 2
2

 ui = K ( + Xi)
2 2 2

ui 2
Ki =
2

( +  Xi) 2
ui 2
K= -----------------------------------------------4.29
( + Xi) 2
ui
K=
 + Xi
The required transformation of equation no (5.12) would be
Y  Xi Ui
= + + ----------5.30
 + Xi  + Xi  + Xi  + Xi

Ui
The new transformed random term is homoscedastic
 + Xi

2
Ui  Ui  1
Var = E   = E (Ui 2 )
 + Xi   + Xi  ( + Xi ) 2

1
ui 2 -----------------------------------------5.31
( + Xi ) 2

Substitute in equation number of (5.27)

1
Var (Ui)= (k 2 ( + Xi) 2 )
( + Xi ) 2

Var (Ui)= K2
Let’s explain using example
Suppose personal saving is dependents upon personal disposable income over a period of 31
year. Given the collected data in the following table 5.1.
S=  + Yd + Ut − − − − − − − − − 4.30
Where S= personal saving, and Yd is disposable income.
Haramaya University 80 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Haramaya University 81 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Table 5
No. Saving Disposabl (Yd Yd )=yd
(S) e (S- S )=s s2 yd 2 yd*s Sˆ = ˆ + ˆYd e=(S- Ŝ ) ei 2
income(Y
d)

1 -
13646.0322 972832.231 186214196. 13459389.7
264 8777 -986.32258 6 8 4 5 95.06082 168.9392 28540.45
2 -
13213.0322 1311763.81 174584221.
105 9210 -1145.32258 6 2 5 15133184.2 131.7186 -26.7186 713.8836
3 -
12469.0322 155476765. 14468099.6
90 9954 -1160.32258 6 1346348.49 5 8 194.7056 -104.706 10963.27
4 -
11915.0322 1252883.03 141967993. 13336764.6
131 10508 -1119.32258 6 8 8 5 241.6073 -110.607 12233.97
5 -
11444.0322 1273111.84 130965874. 12912560.0
122 10979 -1128.32258 6 5 4 1 281.4821 -159.482 25434.55
6 -
10511.0322 1307186.52 110481799. 12017500.5
107 11912 -1143.32258 6 2 2 2 360.4699 -253.47 64247
7 712880.619 8169692.52
406 12747 -844.32258 -9676.03226 1 93625600.3 2 431.161 -25.161 633.0769
8 558491.038 79638351.7 6669130.81
503 13499 -747.32258 -8924.03226 6 8 3 494.8253 8.17466 66.82507
9 671289.490 6680782.74
431 14269 -819.32258 -8154.03226 1 66488242.1 9 560.0135 -129.014 16644.49
10 47624246.2 4570709.49
588 15522 -662.32258 -6901.03226 438671.2 5 1 666.0925 -78.0925 6098.442
11 898 16730 -352.32258 -5693.03226 124131.200 32410616.3 2005783.81 768.3618 129.6382 16806.06

Haramaya University Distance

82 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

4 1 4
12 90193.6520 22657907.1 1429545.16
950 17663 -300.32258 -4760.03226 6 2 9 847.3496 102.6504 10537.11
13 222144.974 14807352.2 1813664.49
779 18575 -471.32258 -3848.03226 4 7 3 924.5595 -145.56 21187.57
14 7773123.88 1202541.26
819 19635 -431.32258 -2788.03226 186039.168 3 8 1014.299 -195.299 38141.74
15 802.168537 1587681.29 35687.3644
1222 21163 -28.32258 -1260.03226 9 6 9 1143.66 78.34042 6137.221
16 204012.491 208819.515 206402.009
1702 22880 451.67742 456.96774 7 4 8 1289.021 412.9792 170551.8
17 107372.491 2903506.05 558351.752
1578 24127 327.67742 1703.96774 6 9 8 1394.592 183.4082 33638.56
18 162955.459 10118555.7
1654 25604 403.67742 3180.96774 4 6 1284084.85 1519.635 134.3654 18054.05
19 22403.3300 16621665.9 610230.012
1400 26500 149.67742 4076.96774 6 5 7 1595.49 -195.49 38216.34
20 334867.556 27530670.4 3036301.75
1829 27670 578.67742 5246.96774 4 6 5 1694.542 134.4578 18078.9
21 901887.202 34538749.8 5581223.56
2200 28300 949.67742 5876.96774 1 2 1 1747.878 452.122 204414.3
22 587794.266 25069725.9 3838729.10
2017 27430 766.67742 5006.96774 3 5 9 1674.224 342.7762 117495.5
23 730473.492 50936308.5 6099805.17
2105 29560 854.67742 7136.96774 3 2 5 1854.55 250.4504 62725.4
24 122274.298 2002591.30
1600 28150 349.67742 5726.96774 1 32798159.5 4 1735.179 -135.179 18273.36
25 999354.944 93643704.6 9673846.14
2250 32100 999.67742 9676.96774 1 4 4 2069.586 180.414 32549.21
26 10076.9677 1368145.26 101545278. 11786801.6
2420 32500 1169.67742 4 7 8 3 2103.45 316.55 100203.9
27 12826.9677 1741548.49 164531101. 16927459.6
2570 35250 1319.67742 4 3 4 9 2336.265 233.735 54632.05

Haramaya University Distance

83 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

28 11076.9677 220596.878 122699214.

1720 33500 469.67742 4 9 3 5202601.63 2188.11 -468.11 219127
29 13576.9677 422080.750 8820649.37
1900 36000 649.67742 4 1 184334053 3 2399.76 -499.76 249760.1
30 13776.9677 721951.718 189804840.
2100 36200 849.67742 4 1 1 11705978.4 2416.692 -316.692 100293.8
31 15776.9677 1101822.68 248912711. 16560726.7
2300 38200 1049.67742 4 6 1 9 2586.012 -286.012 81802.86
Sum 20218310.7 217800819.
38760 695114 7 2572501037 7 -0.35124 1778203
mea 1250.3 22423.032
n 2258 26

Haramaya University Distance

84 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Lower values of the disposable income Table 5.2

N.o Saving Disposable (Yd Yd )=yd
S income(Yd) (S- s2 yd 2 yd*s Sˆ = ˆ + ˆYd e=(S- Ŝ ) ei 2
S )=s
1 264 8777 -67.363 -3414.54 4537.773769 11659083.41 230013.658 32.6414 231.3586 53526.8
2 -
105 9210 226.363 -2981.54 51240.20777 8889580.772 674910.339 70.832 34.168 1167.452
3 -
90 9954 241.363 -2237.54 58256.09777 5006585.252 540059.367 136.4528 -46.4528 2157.863
4 -
131 10508 200.363 -1683.54 40145.33177 2834306.932 337319.125 185.3156 -54.3156 2950.184
5 -
122 10979 209.363 -1212.54 43832.86577 1470253.252 253861.012 226.8578 -104.858 10995.16
6 -
107 11912 224.363 -279.54 50338.75577 78142.6116 62718.43302 309.1484 -202.148 40863.98
7 406 12747 74.637 555.46 5570.681769 308535.8116 41457.86802 382.7954 23.2046 538.4535
8 503 13499 171.637 1307.46 29459.25977 1709451.652 224408.512 449.1218 53.8782 2902.86
9 431 14269 99.637 2077.46 9927.531769 4315840.052 206991.882 517.0358 -86.0358 7402.159
10 588 15522 256.637 3330.46 65862.54977 11091963.81 854719.263 627.5504 -39.5504 1564.234
11 898 16730 566.637 4538.46 321077.4898 20597619.17 2571659.359 734.096 163.904 26864.52
Sum 3645 134107 0.007 0.06 680248.5455 67961362.73 5998118.818 -26.8474 150933.7
mean 331.36364 12191.5455

Haramaya University Distance

85 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Large values of the disposable income Table 5.3

N.o Saving Disposable (Yd Yd )=yd
S income(Yd) (S- s2 yd 2 yd*s Sˆ = ˆ + ˆYd e=(S- ei 2
S )=s Ŝ )
1 - -
1829 27670 261.363 -4823.63 68310.61777 23267406.38 1260718.408 1936.483 107.483 11552.6
2 - -
1600 28150 490.363 -4343.63 240455.8718 18867121.58 2129955.438 1951.795 351.795 123759.7
3 -
2200 28300 109.637 -4193.63 12020.27177 17586532.58 459777.0123 1956.58 243.42 59253.3
4 -
2105 29560 14.637 -2933.63 214.241769 8606184.977 42939.54231 1996.774 108.226 11712.87
5 -
2250 32100 159.637 -393.63 25483.97177 154944.5769 62837.91231 2077.8 172.2 29652.84
6 2420 32500 329.637 6.37 108660.5518 40.5769 2099.78769 2090.56 329.44 108530.7
7 - -
1720 33500 370.363 1006.37 137168.7518 1012780.577 372722.2123 2122.46 -402.46 161974.1
8 2570 35250 479.637 2756.37 230051.6518 7597575.577 1322057.038 2178.285 391.715 153440.6
9 - -
1900 36000 190.363 3506.37 36238.07177 12294630.58 667483.1123 2202.21 -302.21 91330.88
10 2100 36200 9.637 3706.37 92.871769 13737178.58 35718.28769 2208.59 -108.59 11791.79
11 2300 38200 209.637 5706.37 43947.67177 32562658.58 1196266.288 2272.39 27.61 762.3121
Sum 22994 357430 0.007 0.07 902644.5455 135687054.546 4341055.455 0.073 763761.7
mean 2090.3636 32493.6364

Haramaya University Distance

86 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Order table 5.1 in assending order of disposable income Table 5.4

Saving(S) Disposable (Yd Yd )=yd
income(Yd) (S- S )=s s2 yd 2 yd*s Sˆ = ˆ + ˆYd e=(S- Ŝ ) ei 2

1 -
264 8777 -986.32258 13646.03226 972832.2318 186214196.4 13459389.75 95.06082 168.9392 28540.45
2 - -
105 9210 1145.32258 13213.03226 1311763.812 174584221.5 15133184.2 131.7186 -26.7186 713.8836
3 - -
90 9954 1160.32258 12469.03226 1346348.49 155476765.5 14468099.68 194.7056 -104.706 10963.27
4 - -
131 10508 1119.32258 11915.03226 1252883.038 141967993.8 13336764.65 241.6073 -110.607 12233.97
5 - -
122 10979 1128.32258 11444.03226 1273111.845 130965874.4 12912560.01 281.4821 -159.482 25434.55
6 - -
107 11912 1143.32258 10511.03226 1307186.522 110481799.2 12017500.52 360.4699 -253.47 64247
7 406 12747 -844.32258 -9676.03226 712880.6191 93625600.3 8169692.522 431.161 -25.161 633.0769
8 503 13499 -747.32258 -8924.03226 558491.0386 79638351.78 6669130.813 494.8253 8.17466 66.82507
9 431 14269 -819.32258 -8154.03226 671289.4901 66488242.1 6680782.749 560.0135 -129.014 16644.49
10 588 15522 -662.32258 -6901.03226 438671.2 47624246.25 4570709.491 666.0925 -78.0925 6098.442
11 898 16730 -352.32258 -5693.03226 124131.2004 32410616.31 2005783.814 768.3618 129.6382 16806.06
12 950 17663 -300.32258 -4760.03226 90193.65206 22657907.12 1429545.169 847.3496 102.6504 10537.11
13 779 18575 -471.32258 -3848.03226 222144.9744 14807352.27 1813664.493 924.5595 -145.56 21187.57
14 819 19635 -431.32258 -2788.03226 186039.168 7773123.883 1202541.268 1014.299 -195.299 38141.74
15 1222 21163 -28.32258 -1260.03226 802.1685379 1587681.296 35687.36449 1143.66 78.34042 6137.221
16 1702 22880 451.67742 456.96774 204012.4917 208819.5154 206402.0098 1289.021 412.9792 170551.8
17 1578 24127 327.67742 1703.96774 107372.4916 2903506.059 558351.7528 1394.592 183.4082 33638.56
18 1654 25604 403.67742 3180.96774 162955.4594 10118555.76 1284084.85 1519.635 134.3654 18054.05
19 1400 26500 149.67742 4076.96774 22403.33006 16621665.95 610230.0127 1595.49 -195.49 38216.34
20 1829 27670 578.67742 5246.96774 334867.5564 27530670.46 3036301.755 1694.542 134.4578 18078.9
21 2200 28300 949.67742 5876.96774 901887.2021 34538749.82 5581223.561 1747.878 452.122 204414.3
22 2017 27430 766.67742 5006.96774 587794.2663 25069725.95 3838729.109 1674.224 342.7762 117495.5
23 2105 29560 854.67742 7136.96774 730473.4923 50936308.52 6099805.175 1854.55 250.4504 62725.4
24 1600 28150 349.67742 5726.96774 122274.2981 32798159.5 2002591.304 1735.179 -135.179 18273.36
25 2250 32100 999.67742 9676.96774 999354.9441 93643704.64 9673846.144 2069.586 180.414 32549.21

Haramaya University Distance

87 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

26 2420 32500 1169.67742 10076.96774 1368145.267 101545278.8 11786801.63 2103.45 316.55 100203.9
27 2570 35250 1319.67742 12826.96774 1741548.493 164531101.4 16927459.69 2336.265 233.735 54632.05
28 1720 33500 469.67742 11076.96774 220596.8789 122699214.3 5202601.63 2188.11 -468.11 219127
29 1900 36000 649.67742 13576.96774 422080.7501 184334053 8820649.373 2399.76 -499.76 249760.1
30 2100 36200 849.67742 13776.96774 721951.7181 189804840.1 11705978.4 2416.692 -316.692 100293.8
31 2300 38200 1049.67742 15776.96774 1101822.686 248912711.1 16560726.79 2586.012 -286.012 81802.86

Haramaya University Distance

88 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

From table 5.1 we can find the following values

S = 1,250.322 s 2
= 20,218,311 Yd 2
= 1815906482
Yd = 22,423.0322  yd 2
= 257,250,1037
Yds = 217,800,819.7  ei 2
= 1778,203 N=31

From equation 5.31 we can estimate  & 

ˆ =
 yds = 217,800,819.7 = 0.08466
 yd 257,250,1037
2

ˆ = S − ˆYd = 1250.322 − 0.08466(22,423) = −648

Sˆ = − 648 + 0.08644Yd

S .E ( ˆ ) =
e 2

=
1,778,203
= 0.004882
(n − 2) yd 2
31 − 1)2572501037

S .E (ˆ ) =
e  x
2 2

= 118.162
(n − 2)n x 2

R 2
= 1.
 ei 2

= 1−
1,778.203
= 0.909
 yi 2
257,2501037
ˆ 0.08466
tˆ = = = 17.34
ˆ
S .E (  ) 0.004882
ˆ − 648
tˆ = = = − 5.483
S .E (ˆ ) 118.162
Sˆ = −648 + 0.08466Yd
s.e. (118.16) (0.00488) R2=0.909
t (-5.485) (17.34) F=300.73

Now test whether there is hetroscedasticity or not in our model. Let’s test hetroscedasticity using
the goldfeld & Qundt test as follows using the table 5.2 up to 5.4.

Goldfeld&Qundat test

Haramaya University 89 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1st order the observation (given in the table 5.4 in ascending order of disposable income (Yd)
2nd omit 9 central values (quarter of the total sample size i.e. 13 x31  9 . Then you will have two
sets of regression equation one which contains small values of Yds (using table5.2) & the large
values of Yds (using 5.3.)
For small values of Yds (from table number 5.2) we will have

S1 = 331.36  s = 680,248.54  syd = 5998118.81

2
1

Yd1 = 12191.54  yd = 67961362.73

2
ei = 150933.7 2

ˆ =
 syd =
5998118.81
= 0.0882
 yd 2
67961362.73
ˆ = S − ˆYd = 331.364 − (0.0882)(12191.54)  −743
If you calculate S .E (ˆ ) = 189.4 & S .E ( ˆ ) = (0.015)
ˆ
Calculated t values will be t = 
0.0882
ˆ = = 5.88
S .E (  ) 0.015
ˆ − 743
t = = = − 3.922
S .E (ˆ ) 189.4
The estimated equation which holds small values of Yd can be written as follows

Sˆ = −743 + 0.882Yd
s.e. (189.4) (0.015)
t(-3.922) (5.88)

R = 1−
2  ei 2
= 0.778
 Si 2
 ei 2
= 150,933.7
Again the estimated for large value of Yds (see in the table 5.4)

S2 = 2090.36
 s 2 = 902,644.54
 syd = 4,34`,055.45
Yd = 32,493.63
2

 yd = 135687054.54
2

e = 763,761.7
2
2

ˆ =
 syd = 0.319
 yd 2

ˆ = s − 0.0319Yd = 1053.47
Again s.e. (ˆ ) = 710.3 s.e. ( ( ˆ ) = 0.025

Haramaya University 90 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Calculated t values will be

ˆ
t = 
0.0319
ˆ = = 1.276
S .E (  ) 0.025
ˆ 1053
t = = = 1.48
S .E (ˆ ) 710.3
We can write the estimated line for large values of Yds.

Ŝ = 1053.47 + 0.0319Yd
s.e. (710.3) (0.025
t (1.276) (1.48)
2
R =0.846
 e22 = 763,761.7
Now to test whether there is hetroscedasticity or not we should undertake Goldfeld Quadat –
using F tes

F test =
 e22
=
763,761.7
 5.06. this is the computed F value
 ei 150933
2

 N −C 
The table value will have V1 =   − k which is the numerator degree of freedom again
 2 
equal to the denominator degree of freedom
N= 31 number of samples C = 9 omitted variables
K is the number of estimated parameters in our case 2 (ˆ & ˆ )
 31 − 9 
V1 =   − 2 = 9 Again V2=9 F9, 9 from the table is equal to 3.18 and compare the
 2 
calculated F* value which is 5.06 with table value of F i.e. 3.18. Since the calculated value is
greater than the table value we reject the null hypothesis (i.e. there is homoscedasticity) & accept
the alternative and then in this model there is hetroscedasticity or the variance of Ui is not
constant.

The spearman test of Hertoscedasticity

To calculate the spearman test which is used to test the existence of Hetroscedasticity 1st give a
rank following the value of Yd. & then giving the rank for e, which you obtain from table
number

Haramaya University 91 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Order the the value of Yd in assending order & following this order the value of ei ( from table
5.1) Table 5.5
disposable Order of Order of
income (Yd) Yd ei ( Yd-ei) = D (Yd-ei)2=D2

1 8777 1 23 22 484
2 9210 2 15 13 169
3 9954 3 13 10 100
4 10508 4 12 8 64
5 10979 5 8 3 9
6 11912 6 5 -1 1
7 12747 7 16 9 81
8 13499 8 17 9 81
9 14269 9 11 2 4
10 15522 10 14 4 16
11 16730 11 20 9 81
12 17663 12 19 7 49
13 18575 13 9 -4 16
14 19635 14 7 -7 49
15 21163 15 18 3 9
16 22880 16 30 14 196
17 24127 17 25 8 64
18 25604 18 21 3 9
19 26500 19 6 -13 169
20 27670 21 22 1 1
21 28300 23 31 8 64
22 27430 20 29 9 81
23 29560 24 27 3 9
24 28150 22 10 -12 144
25 32100 25 24 -1 1
26 32500 26 28 2 4
27 35250 28 26 -2 4
28 33500 27 2 -25 625
29 36000 29 1 -28 784
30 36200 30 3 -27 729
31 38200 31 4 -27 729
Sum 4826

D
2
r = 1-
n(n 2 − 1)

Haramaya University 92 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1-- 4826 /29760 = 0.837

Since the Spearman rank correlation is high it indicates the existence of hetroscedasticity

5.2.1.4. How to remove Hetroscedasticity

Using table 5.1 we detect hetroscedasticity but the question is how to remove hetroscedasticity.
Let’s assume that the pattern of hetroscedasticity is
ui 2 = k 2 Y 2
ui 2 ui
So that k 2 = 2
=k =
Yd Yd
To remove hetroscedasticity transform the original model by dividing X. Then our saving
equation will be transformed as follows

S 1 Ut
= + +
Yd Yd Yd
Take table number 5.1 and change it as follows
& it is presented in table number 5.6.

Haramaya University 93 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

N.o. Saving Disposable

( S) (Yd) S/Yd I/Yd (S/Yd)2 i /Yd 2 1/yd*s/yd e=(S/yd- Sˆ / yd ) ei 2
1 264 8777 30.0786 1.1393 349.0196 0.3532 -11.1035 -722.4 521858.3
2 105 9210 11.4007 1.0858 1395.7717 0.2924 -20.2034 -722.4 521865.2
3 90 9954 9.0416 1.0046 1577.6060 0.2113 -18.2557 -722.4 521875.8
4 131 10508 12.4667 0.9517 1317.2536 0.1654 -14.7592 -722.4 521882.6
5 122 10979 11.1121 0.9108 1417.4138 0.1338 -13.7730 -722.4 521887.9
6 107 11912 8.9825 0.8395 1582.3005 0.0867 -11.7142 -722.4 521897.2
7 406 12747 31.8506 0.7845 285.9497 0.0574 -4.0499 -722.4 521904.3
8 503 13499 37.2620 0.7408 132.2192 0.0383 -2.2514 -722.4 521910.0
9 431 14269 30.2053 0.7008 344.3006 0.0243 -2.8913 -722.4 521915.2
10 588 15522 37.8817 0.6442 118.3519 0.0098 -1.0797 -722.4 521922.6
11 898 16730 53.6760 0.5977 24.1607 0.0028 0.2592 -722.4 521928.6
12 950 17663 53.7847 0.5662 25.2413 0.0004 0.1063 -722.4 521932.7
13 779 18575 41.9381 0.5384 46.5478 0.0000 0.0453 -722.5 521936.3
14 819 19635 41.7112 0.5093 49.6947 0.0013 0.2517 -722.5 521940.1
15 1222 21163 57.7423 0.4725 80.6692 0.0053 -0.6510 -722.5 521944.9
16 1702 22880 74.3881 0.4371 656.7653 0.0117 -2.7661 -722.5 521949.5
17 1578 24127 65.4039 0.4145 276.9969 0.0170 -2.1724 -722.5 521952.4
18 1654 25604 64.5993 0.3906 250.8613 0.0239 -2.4461 -722.5 521955.5
19 1400 26500 52.8302 0.3774 16.5609 0.0281 -0.6822 -722.5 521957.2
20 1829 27670 66.1005 0.3614 300.6683 0.0337 -3.1835 -722.5 521959.3
21 2200 28300 77.7385 0.3534 839.7150 0.0367 -5.5534 -722.5 521960.3
22 2017 27430 73.5326 0.3646 613.6494 0.0326 -4.4697 -722.5 521958.9
23 2105 29560 71.2111 0.3383 504.0212 0.0427 -4.6406 -722.5 521962.3
24 1600 28150 56.8384 0.3552 65.2490 0.0360 -1.5328 -722.5 521960.1
25 2250 32100 70.0935 0.3115 455.0874 0.0545 -4.9806 -722.5 521965.8
26 2420 32500 74.4615 0.3077 660.5341 0.0563 -6.0990 -722.5 521966.3
27 2570 35250 72.9078 0.2837 583.0835 0.0683 -6.3099 -722.5 521969.4
28 1720 33500 51.3433 0.2985 6.6698 0.0608 -0.6366 -722.5 521967.5
29 1900 36000 52.7778 0.2778 16.1371 0.0714 -1.0735 -722.5 521970.2
30 2100 36200 58.0110 0.2762 85.5693 0.0722 -2.4861 -722.5 521970.4
31 2300 38200 60.2094 0.2618 131.0737 0.0802 -3.2425 -722.5 521972.2
Sum 16.8957 2.1086 -152.3450 16179999.0

Haramaya University Distance

94 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

 1
yd 2 = 2.1085x10 −8  s
yd 2 = 4.49 x10 −10

 1 yd s
yd
= −1,52345x10−5

 s yd . 1 yd
ˆ = = −722.50
 yd1 2

ˆ = s yd − ˆ 1 yd = 0.088

sˆ = −722.50 + 0.88 1
Yd
Sˆt
= ˆ 1 + ˆ
Ydt ydt
Sˆt
= −722.5 1 + 0.088 R2= 0.77
Ydt xt
The transformed equation must be multiplied by Yd & the equation will be
Sˆt = −722.5 + 0.088Ydt. This equation is free of hetroscedasticity

5.2.2. Autocorrelation
One of the assumption of OLS is the successive values of the random term Ui are temporarily
independent i.e. the value of Ui at time t is not correlated with Ui at t-1 period. This is called Ui's
are not- correlated with each other or there is no autocorrelation or Ut &Ut-1 are not serially
dependent (they are serially independent)
This assumption of no autocorrelation (serially independent) states that the covariance between
Ut&Ut-1 is equal to zero or the successive values of Ut & Ut-1 covariance is zero
Cov (UtUt-1) = E{[Ut-E(Ut)][Ut-1-E(Ut-1]}
We know that from our assumption
E(Ut)=0 & E(Ut-1) = 0 then we left E[Ut-Ut-1]=0
But if the assumption is violated i.e. if the value of Ut (the random term at time t) is correlated
with (depend up on) its own previous value (i.e Ut-1) we say that there is autocorrelation or the
random term Ui is serially dependent. Autocorrelation is a special type of correlation & it refers
to only the successive value of the same variable but it doesn't refer to the successive values of
different variables.

5.2.2.1. Graphic methods of detecting autocorrelation

There are two methods that are commonly used to obtain a rough idea of the existence or
absence of autocorrelation in the disturbance term- Ui- since ei = Yi − Yˆi estimates of the true Ui,
and then if e's are found to be correlated it will suggest that Ui are auto correlated with each
others. Then we can use the following methods.

Haramaya University 95 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

By plotting the scatter diagram of ei's i.e the variable correlation we attempt to detect are et & et-
1 the observation point to be plotted are (e1,e2) (e2,e3), (e3, e4)--- (en,en-1) . Suppose if you have
Yi =  + Xi + Ui model
1st estimates the model using the data & find Yˆ = ˆ + ˆXi then find the residual values
i.e ei = Yi − Yˆi . After the value of ei's are found we can calculate the value of et-1, as follows.
et et-1
e2 e1
e3 e2
34 e3
. .
. .
. .
en en-1

In the first case when time period is 2 then et will be e2 & et-1 will be e2-1=e1 by doing so you can
get the values e1e2, e2e3, e3e4 etc. Now plotted these corresponding values on the two dimensional
diagrams. If on plotting, most of the points (etet-1) fall in 1st & 3rd quadrant (as shown in fig a)
we can say that there is positive autocorrelations. i.e. the product between et & et-1 are positive.
If most of the points are fall (etet-1) in quadrant 2nd &4th there will be negative autocorrelation
because the product et et-1 are negative (as shown in fig b).

et et

et-1

et-1

et-1 et-1

et et

Haramaya University 96 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Fig(a) positive outcome Fig. (b) Negative outcome

utocorrelation may be positive or negative but in most of the cases of practice autocorrelation is
positive. The main reason for this is economic variables are moving in the same direction. Ex. in
period of boom employment, investment, output, growth of GNP,consumption etc are moving up
wards & then the random term Ui will follow the same pattern and again in periods of recession
all the economic variables will move down words & the random term will follow the same
patterns.
Another methods commonly used in applied econometric research for the detection of
autocorrelation is to plot the residuals against time (t)- & we will have two alternatives.
If the sign of successive values of the residuals etet-1 are changing rapidly their sign we can say
there is negative autocorrelation.
If the sign of successive values of etet-1 do not change its sign frequently i.e. several positives
are followed by several negatives values of etet-1 we can conclude that there is positive
autocorrelation. This can be seen using the following diagram

+ve et +ve et

Negative e6
autocorrelation e6
e1 e3 e5 e6
e6
e5
e6
e6 e4
t
e3
e2
e2 e4 e6 Positive
e1 autocorrelation

-ve et
-ve et
Other methods of detecting autocorrelations
The Runs Test:- In this test we take in to considerations the signs of residuals i.e. the residuals
will be listed as positive & negative ones in sequence.
Ex. we may have the following sequences
(-e1,-e2,-e3,-e4,-e5,-e6,-e7,-e8) (+e9, +e10, +e11, +e12, +e13, +e14, +e15, +e16, +e17, +e18, +e19, +e20,
+e21,) (-e22) (+e3) (-e24,-e25,-e26,-e27,-e28,-e29,-e30,-e31,-e32)

Haramaya University 97 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Thus there are 8 negative residuals followed by 13 positive residuals, followed by a negative &
a positive residuals, and finally followed by 13 negative residuals.
We now define a run as uninterrupted sequence of one symbol or attribute such as -ve or +ve
Length of run = as the number of elements in the run. From our above sequence of e's we have 5
runs i.e. uninterrupted values of e's i.e. 8-ve e's followed by 13 +ve e's, 1 -ve e, by 1 +ve & then
by 13 -ve e's. By examining how run behave in strictly random sequences of observation one can
derive a test of randomness of runs. If there are too many runs, it would mean that in our
example the e's are changing the signs frequently, thus indicating negative autocorrelation.
Similarly if there are too few runs, they may suggest positive autocorrelation.
Now let’s represent the variables
n= total number of sample observations, equal to n1 +n2
Where n1 is the number of +ve symbols of e's
n2 is the number of -ve symbols of e's
K= number of runs
Assuming that n1>10 (the values of e's) & n2>10 (-ve values of e's) the number of runs is
distributed normally with
2n1 n2
Mean = E(K) = +1
n1 + n2
2n n ( 2n n − n − n )
Variance of K =  2 k = 1 2 12 2 1 2
(n1 + tn2 ) (n1 + n2 − 1)
Now using our hypothesis testing we hypothesis as follow
The null hypothesis H0= Cov (etet-1) = 0. No autocorrelation against
The alternative H1 Cov (etet-1)  0. There is autocorrelation
There is autocorrelation and the establish confidence interval with 95% confidence.
E ( x) − 1.96k  k  E ( K ) + 1.96k 
If the estimated k is lies in this limit accept the null hypothesis that there is no autocorrelation &
reject the existence of autocorrelation. Again if the estimated K is lies outside this limit reject the
null hypothesis & accept the alternative that there is autocorrelation.
In our example
n1=14 i.e. total number of +ve residuals (e's)
n2 = 18 i.e total number of -ve residuals (e's)
Using our formula
2(14)(18)
E(K ) = + 1 = 16.75 Mean value of K
14 + 8
2(14)(18)2(14)(18) − 14 − 18
Var(K) =  2 k = = 7.49395
(14 + 18) 2 + (14 + 18 − 1)
S .E (k ) =  2 = 7.49395 = 2.7375
The 95% confidence interval is calculated as follows. From 95% confidence interval the level of
significance is 1- 0.95 which is equal to 0.05 and again we will have two tail test then
 = 0.05 = 0.025 from the normal table 0.025 value will be obtained at the intersection point
2 2
of 1.9 in the first column & 0.06 in the first raw of the normal table. Then the normal distribution
value of the two tail test of 5% significance level is 1.96. Hence the 95% confidence interval of
k( number of Runs)

Haramaya University 98 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

K  1.96k & this will be

16.75  1.96(2.7375) = 11.3845 & 22.11.5
Or this can be written as
11.3845  k  22.1155
But the calculated run value (K) is 5 form our example i.e. 1st -ve e's, 2nd +ve e's 3rd -ve's, 4th
+ve's 5th - ve's. Since the calculated value of K is lying outside the table value of K we reject the
null hypothesis that ther is no autocorrelation between e’s & accept the alternatives that e's are
correlated with each other.
Von Neumen ratio tests for the existence of autocorrelation. This is the ratio of the variance of
the first difference of any variable X over the variance of X. And this test is applicable for
directly observed series & for variables which are random i.e. variables whose successive values
are not auto correlated.
n

 2  ( xt − xt − 1) 2
n −1
= i =2

Sx 2  ( xt − x ) 2

n
In case of the random term Ui this values are not directly observable but are estimated from the
OLS residuals (e's). For large samples (n>30) the Neuman ratio is
n

 (et − et − 1)
i=2
2

2 n −1
=
Se 2
 ( xt − x ) 2

n
Since the method is applicable only for large sample size i/e/ n>30. Then as the sample size is
2
very large then may be approximately normally distributed with
Se 2
 2  2n
E  2  =
 Se  n − 1
 2  4n 2 (n − 2)
Variance = Var  2  =
 Se  (n + 1)(n − 1)
2

To under take the test 1st compute the Von Numan ration 2ndly by using the formula for mean &
variance determine the confidence interval. Then
if the calculated value of Von Numan ratio is lies in the confidence interval accept the null
hypothesis that there is no autocorrelation reject the alternative.
If the calculated value of Neuman test ratio is lie outside the confidence interval we reject the
null hypothesis & accept the alternative that there is autocorrelation in the model.
Durbin - Watson Test
The most celebrated test for the defection of serial correlation (autocorrelation) is developed by
two statisticians Durbin & Watson. It is popularly known as the Durbin- Watson d- statistics.
This test is applicable
For small samples
This is based upon the sum of the squared differences in successive values of the estimated
disturbance term.

Haramaya University 99 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

The test is appropriate only for the first-order autoregressive scheme i.e. only when the
successive values are correlated with each other meansif et is correlated with et-1 et etc. but not
et correlated with et-2 & so on.
The d- statistics will be obtained using the following formula.
n

 (et − et − 1) 2

d= t =1
n

 et
t =1
2

From the above formula we can observe the following situation

If the successive values of the disturbance term is tend to close to another i.e. if positive value of
et is followed by positive value of et-1. The value in the numerator (et-et-1)2 will be very small.
Then we would expect positive autocorrelation results in small values of d.
If there is large differences between et & et-1 i.e when the successive values of et & et-1 have
different signs it will generate large difference in the numerators. The signal for these types of
autocorrelation is an unusually large value of d which indicates negative autocorrelation.
The d- statistics test may be outlined as follows.
Step 1: If the null hypothesis H0 = p= 0 means that there is no- auto correlation (the e's are
serially independent or e's are not serially dependent). Against the alternative that H1=P=0
means that the e's are serially correlated (dependent) with each other or there is autocorrelation
between the successive values of et & et-1. Then we compute the d- statistics to test the null
hypothesis
n

 (et − et − 1) 2

d= t =2
n

 et
t =1
2

And this
n
(et 2 − 2etet − 1 + et 2 − 1
d = n
t =2
 et
t =1
2

n
(et − 2etet − 1 + et 2 − 1
2
d = n
t =2
 et
t =1
2

n n n

 et 2 +  et 2 − 2 etet − 1
d= t =2 t =2
n
n=2

 et
t =1
2

If n is very large ( if we have large sample size) the following terms

n n n

 et 2 
t =2
 et 2 − 1 
t =2
 et
t =1
2
are approximately equal there for we may write

 2 et 2 − 1 2 etet − 1
d = − 
  et − 1  e2t − 1 
2

Haramaya University 100 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

  et 2 − 1  etet − 1
d = 2 − 
  et − 1  e t − 1 
2 2

  etet − 1
d = 21 − 
  e t − 1 
2

Let Pˆ =
 etet − 1 (We will prove this latter on)
 et 2 − 1
d  2(1 − Pˆ )
From the above relations we will have the following
if there is no autocorrelation i.e. Pˆ = 0 the d = 2
if Pˆ = 1 then d-statistics will be zero ( d=0) & we will have perfect positive autocorrelation
If Pˆ = −1 d will be 4 then there will perfect negative autocorrelations.
Now we can conclude from the above values of P̂ the value of d is lies between 0&4 (o<d<4).
The next step is to compare the computed value of d* with the table value of d. The problem
associated with this test is the exact distribution of d is not known. Due to this Durbin & Watson
have established arange of values with in which we can neither accept nor reject the null
hypothesis. These are the upper (dU) & lower limits (dL) for the significance level of d which
are appropriate to test no autocorrelation against the alternative hypothesis of the existence of
autocorrelation. For the two tailed Duration- Watson test we have set of five regions for the
value of d in diagram as follows.
Inconclusive region
H0

Critical region
Inconclusive region

Reject H0& accept

that there is –ve
autocorrelation
Critical region

Accept H0 autocorrelation
No
d
Reject

autocorrelation
dL du (4-du) (4.dL)

If calculated d is less than dL (i.e. d<dL ) we reject the null hypothesis & accept the alternative
that there is autocorrelation & the type of autocorrelation is positive autocorrelation
If the calculated d is less than (4-dL) i.e [d<(4-dL)] we reject the null hypothesis of no
autocorrelation & accept the alternative that there is autocorrelation. The type of autocorrelation
is negative autocorrelation.
If the calculated d value is lieing between du & 4-du [du<d<(4-du)] we accept the null
hypothesis of no autocorrelation & reject the alternative.
If the calculated d have the following type

Haramaya University 101 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

[dL<d<du] of if (4—du)<d<(4-du)<d<(4-dL) the test will be inconclusive.

Short comings of the Durbin-Watson test

If the regression equation has no intercept term i.e if Yi = Xi + Ui in this case the regression is
assumed to pass through the origin. Then the econometrician should re-run the regression by
including the intercept term to obtain e’s.
The distribution of U’s are generated by the first order auto regressive means et is depend upon
et-1, & et-1 on et-2 & so on but it will not be applicable if et is depends upon et-2.
The regression equation does not include the lagged values of one of the explanatory variables.

An alternative test for Autocorrelation

This alternative test for autocorrelation has the advantage of applicability to any form of
autocorrelation & it provides estimates of the coefficients of the autocorrelation relationship,
which are required for remedial transformation of the original observations. The procedure is as
follows.
1st apply OLS to the sample observation i.e

Yi =  + Xi + U
Regress your model by fitting the data & obtain the values of the estimated residuals i.e. e’s
2nd Since we are not sure in a priori about the existence of autocorrelation, we may experiment
with various forms of autoregressive structures, for example

et= Pet-1 +vt

Or et=Pet2-1 +vt or
Or et=P1et-1+P2et-2 +vt or
et = P et − 1 + vt and so on
Vt satisfies all the assumption of OLS. In the above equation et represents (Yt − Yˆt ) values & et-
1 represents all one year lagged values of et. p is the estimated coefficients (estimator) Then the
statistical significance of the estimation P̂ & the over all significance of the above fitted
regression will be judged on the basis of S.E test, t test & F test. If P̂ is found statistically
significant means

Pˆ Pˆ
 S .E ( Pˆ ), and , t 
2 S .E ( Pˆ )
R2
F= k −1  F . This implies that the calculated value (F*) is greater than the table
(1 − R ) 2

N − k)
value (F) then we accepts that there is autocorrelation in the model

Haramaya University 102 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

5.2.2.2. Consequence of autocorrelation

An OLS technique is based on the basic assumptions of mean, variance & covariance of the
random term Ui. If these assumptions do not hold well on what so ever account, the estimator’s
derived using OLS procedure may not be efficient.

OLS estimates are unbiased

Even after the residuals are serially correlated the parameter estimates of OLS are statistically
unbiased. i.e their expected value is equal to the true parameters.


 =  +  xi Ui
 xi 2
1 1

We have already formulated the above.

Taking the expected values

E ( ˆ ) = E ( 1 ) +
 xi (Ui)
 xi 2
We know that E(Ui)=0 then
ˆ) =
E (  +0

E ( ˆ ) = 
Thus irrespective of whether the random term ei is serially independent or not the estimates of
the parameter ( ˆ ) have any statistical bias as long as Ui& Xi’s are uncorrelated.

The variances of OLS estimates are underestimated.

When there is autocorrelation in the random term Ui the variance of the parameter estimates are
likely to be larger than those parameters estimated using other econometric methods.

( ˆ ) = 1 +
 xiui
 xi 2
Var ( ˆ ) = E[ ˆ − E ( ˆ )]2
1

We know that E ( ˆ ) = 

& 1 =  1 +
 xiUi then substitute these values in Var ( ˆ )
 xi2
 xiUi −  
2

Var ( ˆ ) = E  1 +
  xi 2 
  xiUi 
2

E 2 
  xi 
2
 xi  xixj
Var ( ˆ ) =   2
E (ui 2 ) + 2 E (uiuj)
  xi  ( xi) 2
From the above formula if Ui’s are serially independent E(UiUj)=0 & the term

Haramaya University 103 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

xixj
2 E (uiuj) = 0 and left with
( xi) 2
2
 xi 
  xi 2  E (ui 2 )
  
When the ui are serially dependent (when there is autocorrelation)
2
 xi  xixj
Var ( ˆ ) =   2
E (ui 2 ) + 2 E (uiuj)
  xi  ( xi) 2
If you compare the two variance of ˆ [for no autocorrelation] & equation of autocorrelation the
variance of ( ˆ ) with no autocorrelation is smaller than the var( ˆ ) with autocorrelation.
The variance of the random term Ui may be seriously underestimated if the U’s are auto
correlated. In particular the var of Ui will be seriously underestimated in the case of positive
autocorrelation of the error term (Ui). As a result R2 ,t & F statistics are exaggerated.
Finally if the value of U is auto correlated, the prediction based up on OLS estimates will be
inefficient. It means the forecasting made on OLS estimates will be incorrect because this
prediction will have large variance as compared with prediction based on estimators obtained
from other techniques.

5.2.2.3. Sources of autocorrelation

The random term Ui may be correlated with preceding random term or autocorrelation may be
observed for many reasons.
Omission of explanatory variables:- Most of economic variables are tend to be auto correlated
with each other if we exclude those explanatory variables which are auto correlated, so their
influence will be captured by the random term Ui. This case may be called “qusi-autocorrelation’
since the autocorrelation is occurred due to the pattern of omitted explanatory variables. In this
case the autocorrelation is occurring not due to the behavior of the random term Ui but following
the pattern of omitted explanatory variable. If several auto correlated explanatory variables are
omitted then their effect on the random term may not be observed because the autocorrelation
pattern of the omitted regressions may be cancelled out with e ach others.
Misspecification of the mathematical model. If we have adopted a mathematical form which
differs from the true form of the relation ship, the U’s may show serial correlation. For example
if we have chosen a linear function.

Yt =  + 1 X 1i +  2 X 2 i + Ui
While the true relationship is

Yt = X 1i 1 X 2 i2 e ui
Which transform in to log Yt = log  + 1 log X 1i +  2 log X 2 i + Ui .
Then if you estimate using the linear functions then in this case Ui will appear to be serially
dependent. But had the researcher uses the log model Ui will not be serially dependent.
Interpolations in the statistical data. In most of the published time series data when some values
of the true disturbance term Ui will be interrelated & exhibit autocorrelation pattern.

Haramaya University 104 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Misspecification of the true random term –Ui. In some cases the time value of Ui may be
correlated successively. For example war which is captured by a random term Ui may affect the
current year production & exerts its influence in some future period production. Or drought in
Ethiopia will affect the crop yield in the agricultural sector which intern influences the
performance of other sectors of the economy in several time periods. Such causes result in
serially dependent autocorrelation (values of the disturbance term Ui). Then during this period of
time if we specify the random term E(UiUj)=0 then we are misspecifying the true pattern of Ui.
This kind of autocorrelation is called “true autocorrelation” because its root lies in the random
term Ui itself.

5.2.2.4. Solution for autocorrelation

The solution for autocorrelation or the type of corrective action in each particular autocorrelation
is depends on the causes or sources of autocorrelation.
If the source is omitted variables then the appropriate measure is to incorporate those excluded
explanatory variables.
Ex. If we have qusi- autocorrelation. Consumption in period t (Ct) is depends upon current
income (Xt), previous income (Xt-1), wealth (W), previous consumption behavior of individual
(Ct-1).
Ct =  + 1 Xt +  2 Xt − 1 + Wt + Ct − 1 + Ut
Now if you omitted the lagged income then its influence will be reflected in the random term Ui
& autocorrelation will occur.
If you omitted lagged income & wealth their effect may be cancelled out with each other & their
influence on the random term may not be observed & Ui will be serially independent.
To eliminate the autocorrelation which appear following the omitted variable (Xt-1) is to
introduce the omitted variables in to the function.
If the source of autocorrelation is misspecification of the mathematical form of the relationship
the relevant approach is to change the functional form. This can be investigated by regress the
residuals against higher powers of the explanatory variables or by computing a linear in logs
form & re-examining the resulting new results.
Given the above cases, if autocorrelation is observed the appropriate procedure is
To transform the original data
Applying OLS to the transformed data

Transformation of the original dataThe transformation of the original data depends up on the
patterns of the autoregressive structure which may be first order or higher order autoregressive.

First- Order autoregressive structure

Suppose we have the model

Y =  + 1 Xt + Ut
If in this model the autocorrelation is the first order scheme means Ut is depend up on Ut-1 i.e.
e1e2, e2e3, e3e4 etc. This is called also first order autoregressive. Then Ut is correlated with its
preceding values.
Ut= Put-1+vt

Haramaya University 105 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Where vt satisfies all the assumption of Ui (OLS assumptions). P= is coefficient of Ut-1 since Ut
is not observable but we can approximate using the sample observation by obtaining et. There
fore
et= Put-1+vt
The estimated value will be
eˆt = Pˆ et − 1
Vt = et − eˆt
 v 2t =  (et − eˆt ) 2
=  (et − pˆ et − 1)
Because eˆt = Pˆ et − 1
vt 2
= −2 (et − pˆ et − 1)(et − 1) = 0
pˆ
etet − 1 − pˆ et 2 − 1 = 0
pˆ =  et 2 =  etet − 1

pˆ =
 etet − 1
 et − 1
2

This is the estimated value of autoregressive of the first order i-e when et is correlated with et-1.
To transform the original data take the lagged form of equation.

Yt − 1 =  + Xt − 1 + Ut − 1
Multiply the above equation by p̂
p̂ Yt − 1 = pˆ  + pˆ Xt − 1 + pˆ Ut − 1
Subtract this equation from the original equation

(Yt − pˆ Yt − 1) = ( − pˆ  ) + (Xt − pˆ Xt − 1) + (Ut − pˆ Ut − 1)

Now let’s represent
Yt − pˆ Yt − 1 = Y *
 − pˆ  =  *
Xt − pˆ Xt − 1 = Xt *
Ut − pˆ Ut − 1 = vt
We can write the above equation as follows.
Y * =  * +X * t + vt
Now we can apply OLS to the transformed relation to obtain ˆ * &ˆ *
 * =  − p̂
* = (1 − pˆ )
*
* =
1− p
 1 
Var (ˆ ) =   var( *)
1 − pˆ 

Haramaya University 106 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

The estimators obtained are efficient, if only our sample is large so that loss of one observation
becomes negligible. The above procedure is possible only when the value of p̂ is known. Now
we can describe the method through which the parameters of the auto correlated model can be
estimated.

Method I. A priori in formation on p̂

In many of the econometric research an investigator make some reasonable guess about the value
of the autoregressive coefficient by using his knowledge of the relationship between variable
under the study. A usual case is to assume that p̂ =1. The model will be transformed as follows.
Yt =  + Xt + Ut
Lag this model by one period then you will have
Yt − 1 =  + Xt + Ut − 1
Multiply by p̂
pˆ Yt − 1 = pˆ  + pˆ Xt − 1 + pˆ Ut − 1
If we assume that p̂ =1 (perfect positive autocorrelation) then we left with
Yt − 1 =  + pˆ Xt − 1 + Ut − 1
Subtract this equation from the original

(Yt − Yt − 1) = ( −  ) +  ( Xt − Xt − 1) + Ut − Ut − 1
Let Yt-Yt-1=Yt*
Xt-Xt-1= X*
Ut-Ut-1=Vt
Then we can write
Yt* = Xt * +vt
Here  is suppressed in this case & we will have the equation that will passes through the origin.
Suppose one assumes that there is perfect negative autocorrelation i.e pˆ = −1
The original model is

Yt =  + Xt + Ut
Lagged the original model by one year.

Yt − 1 =  + Xt − 1 + Ut − 1
pˆ = −1 Multiply by (-1) the lagged value
− Yt − 1 = − − Xt − 1 − Ut − 1 Subtract from the original model
Yt − (−Yt − 1) =  − ( ) + Xt − (−Xt − 1) + Ut − (−Ut − 1)
Yt + Yt − 1 = 2 +  ( Xt + Xt − 1) + Ut + Ut − 1
Divided both sides by 2 (because to make free the intercept term  )
Yt + Yt − 1 ( Xt + Xt − 1 Ut + Ut − 1
= + )+
2 2 2
This model is called two period moving average regression models because we are regressing the
value of one moving average
Yt + Yt − 1 Xt + Xt − 1
on
2 2

Haramaya University 107 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

This method of first difference is quite popular in applied research for its simplicity. The
problem with this method is that it depends up on the assumptions of the perfect positive or
perfect negative autocorrelation in the data. But now the question arises how to know whether
the value of p̂ is equal to +ve or –ve 1? The answer is estimation of p̂ in the following ways.

Method II. Estimation of p̂ from the d* statistics

Another crude method for the estimation of the coefficient of the autoregressive scheme is to
solve for the p̂ .
d  2(1 − pˆ )
Suppose we calculate certain value of d- statistics
1
pˆ = 1 − d *
2
p̂ Will not be accurate if the sample size is small. The above relationship of d-statistics will be
true for large samples. For small samples Theil & Nagar have suggested the following relation
n 2 (1 − d ) + k 2
pˆ = 2
n2 − k 2

Where n= total number of observations

d= Durban – Watson statistics
k= number of coefficients (including intercepts) to be estimated
Example - Given the following model Yt =  + Xt + Ut

Table 5.7
Y X (X- x )2=x2 (Y- Y )2=y2 xy Ŷ = ̂ + ˆ X (Y- Ŷ )=e e2
1 2 1 49 25 35 0.63 1.37 1.8769
2 2 2 36 25 30 1.54 0.46 0.2116
3 2 3 25 25 25 2.45 -0.45 0.2025
4 1 4 16 36 24 3.36 -2.36 5.5696
5 3 5 9 16 12 4.27 -1.27 1.6129
6 5 6 4 4 4 5.18 -0.18 0.0324
7 6 7 1 1 1 6.09 -0.09 0.0081
8 6 8 0 1 0 7 -1 1
9 10 9 1 9 3 7.91 2.09 4.3681
10 10 10 4 9 6 8.82 1.18 1.3924
11 10 11 9 9 9 9.73 0.27 0.0729
12 12 12 16 25 20 10.64 1.36 1.8496
13 15 13 25 64 40 11.55 3.45 11.9025
14 10 14 36 9 18 12.46 -2.46 6.0516
15 11 15 49 16 28 13.37 -2.37 5.6169
Sun 105 120 280 274 255 41.768
Average 7 8

And given data for Y & X in table 5.7

Haramaya University 108 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

We can get the following values

Y =7
x =8
 yi2 = 274
 xi = 280
2

 xiyi = 255
ei = 41.76874
2

ˆ =
 xiyi = 255 = 0.91
 xi 280 2

ˆ = y − ˆx = 7 − (0.91)(8) = 7 − 7.28 = −0.28

r 2 = 1
 xiyi = 0.91 (255) = 0.85
 yi 2
274
Yˆ = −0.28 + 0.91Xt

S .E ( ˆ ) =
 ei 2
1
. =
41.768
= 0.107
n − k  xi 2 3640

n=14 & K=2( ˆ & ˆ )

S .E (ˆ ) =
 ei .  xi
2 2

=
41.768 x1240
=
51792
n − k n xi 2
(13)(15)(280 54600
S .E (ˆ ) = 0.948

Derived from table number 5.7 table 5.8

Et et-1 et2 et—et-1 (et--et-1)2 (et-1)2
1 1.37 ---- 1.8769 1.37 1.8769 0
2 0.46 1.37 0.2116 -0.91 0.8281 1.8769
3 -0.45 0.46 0.2025 -0.91 0.8281 0.2116
4 -2.36 -0.45 5.5696 -1.91 3.6481 0.2025
5 -1.27 -2.36 1.6129 1.09 1.1881 5.5696
6 -0.18 -1.27 0.0324 1.09 1.1881 1.6129
7 -0.09 -0.18 0.0081 0.09 0.0081 0.0324
8 -1 -0.09 1 -0.91 0.8281 0.0081
9 2.09 -1 4.3681 3.09 9.5481 1
10 1.18 2.09 1.3924 -0.91 0.8281 4.3681
11 0.27 1.18 0.0729 -0.91 0.8281 1.3924
12 1.36 0.27 1.8496 1.09 1.1881 0.0729
13 3.45 1.36 11.9025 2.09 4.3681 1.8496
14 -2.46 3.45 6.0516 -5.91 34.9281 11.9025
15 -2.37 -2.46 5.6169 0.09 0.0081 6.0516

Haramaya University 109 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

sum 41.768 60.2134 36.1511

Test the existence of Durbin-Watson test

d* =
 (et.et − 1) 2
 et 2
et = 41.768
2

 (et − et − 1) = 60.2134
2

60.2134
d* = = 1.442 . This is the calculated value of d. The table value of d at 5% significance
41.768
level N=15 & one explanatory variable (k=1) is DL= 1.08 du = 1.36. Since D* (calculated 1.44)
is greater than du (1.36) and again d* is less than (4-du) i.e. 2.64 the calculated statistics is lying
between du & 4-du then there is no autocorrelation in the given sample of X & Y

F(d)
No autocorrelation
1.36=dU

2.64=4-dU

Example 2
1.08=dL

Suppose the hypothetical data on consumption expenditure and disposable income are given in
the table 5.9 if the estimated function is given by
C =  + Xt + Ut
Where C= consumption expenditure
Yt= disposable income

Haramaya University 110 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Haramaya University 111 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

S.n Ct Yt C Y c2 y2 Cy ĉ (C- et2 et-1 (et-et- (et-et- (et-1)2 et*et-

ĉ )=et 1) 1) 1
1 -
206.3 226.5 -140.647 152.84 19781.69 23358.84 21495.99 208.48 -2.18 4.75 0.00 0.00
2 -
216.7 238.6 -130.247 140.74 16964.39 19806.62 18330.50 219.44 -2.74 7.52 -2.18 -0.56 0.32 4.75 5.97
3 -
230 252.6 -116.947 126.74 13676.69 16062.01 14821.45 232.13 -2.13 4.52 -2.74 0.62 0.38 7.52 5.83
4 -
236.5 257.4 -110.447 121.94 12198.63 14868.39 13467.51 236.47 0.03 0.00 -2.13 2.15 4.63 4.52 -0.05
5 -
254.4 275.3 -92.5474 104.04 8565.02 10823.49 9628.26 252.69 1.71 2.92 0.03 1.68 2.83 0.00 0.04
6 266.7 293.2 -80.2474 -86.14 6439.65 7419.41 6912.19 268.91 -2.21 4.88 1.71 -3.92 15.35 2.92 -3.77
7 281.4 308.5 -65.5474 -70.84 4296.46 5017.74 4643.12 282.77 -1.37 1.88 -2.21 0.84 0.70 4.88 3.03
8 290.1 318.8 -56.8474 -60.54 3231.63 3664.61 3441.31 292.10 -2.00 4.01 -1.37 -0.63 0.40 1.88 2.75
9 311.2 337.3 -35.7474 -42.04 1277.88 1767.03 1502.68 308.86 2.34 5.46 -2.00 4.34 18.83 4.01 -4.68
10 325.2 350 -21.7474 -29.34 472.95 860.60 637.98 320.37 4.83 23.33 2.34 2.49 6.22 5.46 11.28
11 335.2 364.4 -11.7474 -14.94 138.00 223.08 175.46 333.42 1.78 3.18 4.83 -3.05 9.28 23.33 8.61
12 355.1 385.5 8.1526 6.16 66.46 37.99 50.25 352.53 2.57 6.59 1.78 0.78 0.61 3.18 4.58
13 375 404.6 28.0526 25.26 786.95 638.27 708.72 369.84 5.16 26.65 2.57 2.60 6.74 6.59 13.25
14 401.2 438.1 54.2526 58.76 2943.34 3453.21 3188.10 400.19 1.01 1.02 5.16 -4.15 17.23 26.65 5.22
15 432.8 473.2 85.8526 93.86 7370.67 8810.45 8058.47 431.99 0.81 0.66 1.01 -0.20 0.04 1.02 0.82
16 466.3 511.9 119.3526 132.56 14245.04 17573.21 15821.86 467.05 -0.75 0.56 0.81 -1.56 2.44 0.66 -0.61
17 492.1 546.3 145.1526 166.96 21069.28 27876.98 24235.26 498.22 -6.12 37.43 -0.75 -5.37 28.80 0.56 4.60
18 536.2 591 189.2526 211.66 35816.55 44801.65 40057.96 538.72 -2.52 6.33 -6.12 3.60 12.97 37.43 15.39
19 579.6 634.2 232.6526 254.86 54127.23 64955.66 59294.77 577.86 1.74 3.04 -2.52 4.26 18.15 6.33 -4.39
Sum 6592 7207.4 223468.51 272019.24 246471.84 144.73 145.92 141.68 67.87

Table 5

Haramaya University Distance

112 Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

From the table we found the following information

 Ct = 6592
Yt = 7387.4
ct = 223,468.5
2

c = 346.95
Y = 379.3368
 yi2 = 272,019.2
ctyt = 246,471.8
ˆ = 
ctyt 246,471.8
= = 0.906
 yt 272,019.2
2

ˆ = c − ˆx = 246.95 − (0.906)(379.3368) = 3.27

cˆ = 3.27 + 0.906Yd
ˆ  ctyt
R2 = = 0.99
 ct 2
Now the estimated consumption function shows that almost all variation in consumption is
explained by disposable income because R2=0.99 (99%). Let’s examine the error terms ui is auto
correlated or not.

d=
 (et − et − 1) 2
 et 2
 (et − et − 1) = 145.916
2

 et = 144.7274
2

145.916
d= = 1.0082 This is the calculated value of d. At 5% significance level n= 19 & K=1
144.7274
the table value is d1=1.18 & du=1.40 since our calculated d-value 1.008 is falls below the lower
bound value of dL, then we can conclude that we reject the null hypothesis of no autocorrelation
& we accept the alternative that there is positive autocorrelation in the disturbance term.

5.27 Solution to remove autocorrelation in our model

Assuming that there exists first order autocorrelation form i.e. e1 is depending upon e2, e2 on e3
and so on. Then
et = Pet-1 +vt
Where vt satisfies all the assumption of Ut.

pˆ =
 etet − 1 From the above data
 et 2 − 1
The following results  etet − 1 = 67.8732 &  et 2 − 1 = 141.683
67.8732
pˆ = = 0.479
141.683
Now we transform the consumption expenditure as follows

Haramaya University 113 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Ct =  + Yt + Ut
Lag by one period
Ct − 1 =  + Yt − 1 + Ut − 1
Multiply it by P̂ which is 0.48 in our case.
0.48Ct − 1 = 0.48 +  0.48Yt − 1 + 0.48Ut − 1.
Subtract it from the original consumption expenditure function & you will have

Ct = 0.48Ct − 1 = (1 − 0.48) + Yt − 0.48Yt − 1) + (Ut − 0.48Ut − 1 ).

Now C * t = Ct − 0.48Ct − 1
Y * t = Yt − 0.48Yt − 1
V * t = Ut − 0.48Ut − 1

First transform the previous data i.e table 5.9 & the new estimated model will be

C * t = a + Y * t + vt
c *2
t =36,975.91 y *2
t = 45,270.8

c yt * = 40,854.01 Y * = 179
2
*
t

C * = 163.7436,975.91

Haramaya University 114 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Table 5.10
sn Ct-1 Yt-1 .48Ct-1 .48Yt-1 C* Y* c* y* c*2 y*2 c*y*
1 -
216.7 238.6 104.016 114.528 102.284 111.972 61.45289 -67.028 3776.458 4492.753 4119.064
2 -
230 252.6 110.4 121.248 106.3 117.352 57.43689 -61.648 3298.996 3800.476 3540.869
3 -
236.5 257.4 113.52 123.552 116.48 129.048 47.25689 -49.952 2233.214 2495.202 2360.576
4 -
254.4 275.3 122.112 132.144 114.388 125.256 49.34889 -53.744 2435.313 2888.418 2652.207
5 -
266.7 293.2 128.016 140.736 126.384 134.564 37.35289 -44.436 1395.238 1974.558 1659.813
6 -
281.4 308.5 135.072 148.08 131.628 145.12 32.10889 -33.88 1030.981 1147.854 1087.849
7 -
290.1 318.8 139.248 153.024 142.152 155.476 21.58489 -23.524 465.9074 553.3786 507.7629
8 -
311.2 337.3 149.376 161.904 140.724 156.896 23.01289 -22.104 529.5931 488.5868 508.6769
9 -
325.2 350 156.096 168 155.104 169.3 8.632889 -9.7 74.52677 94.09 83.73902
10 -
335.2 364.4 160.896 174.912 164.304 175.088 0.567111 -3.912 0.321615 15.30374 2.218539
11 355.1 385.5 170.448 185.04 164.752 179.36 1.015111 0.36 1.030451 0.1296 0.36544
12 375 404.6 180 194.208 175.1 191.292 11.36311 12.292 129.1203 151.0933 139.6754
13 401.2 438.1 192.576 210.288 182.424 194.312 18.68711 15.312 349.2081 234.4573 286.137
14 432.8 473.2 207.744 227.136 193.456 210.964 29.71911 31.964 883.2256 1021.697 949.9417
15 466.3 511.9 223.824 245.712 208.976 227.488 45.23911 48.488 2046.577 2351.086 2193.554
16 492.1 546.3 236.208 262.224 230.092 249.676 66.35511 70.676 4403.001 4995.097 4689.714
17 536.2 591 257.376 283.68 234.724 262.62 70.98711 83.62 5039.17 6992.304 5935.942
18 579.6 634.2 278.208 304.416 257.992 286.584 94.25511 107.584 8884.026 11574.32 10140.34
sum 2947.264 3222.368 -2E-07 0.368 36975.91 45270.8 40854.01

Where C*=Ct-.48Ct-1, Y*=Yt-.48Yt-1,

c* = C*- C * y*= Y*- Y *

ˆ =  c y t = 40,854.01 = 1.10488
* *
t t

 c t 31,975.91
* 2

aˆ = C * ˆY * t = 163.74-(1.10488)(179.02)= -34.9722

aˆ − 34.9722
ˆ = = = 67.254
ˆ
1 − P 1 − 0.488

The regression model after the data is transformed can be written as

Cˆ * = −34.9722 + 1.11Yt *
This model after transformation of the data can be expressed in terms of original variable as

Cˆ t = −34.9722 + 1.11Yt.
n=18

Haramaya University 115 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Example 3: Given the following model

Model A = Yt =  + 1t + ut
Model B= Yt =  + 1t +  2 t 2 + ut
Where t is time then regression on data for 1948-1964 gave the following results.

Model A = Yˆ = 0.4529 − 0.041t

(-3.9608) R2=0.5284 d=0.8252
ˆ
Model B= Yt = 0.4786 − 0.0127t + 0.0005t 2
t=(-3.2724) (2.777)
R2=0.6629
d=1.82
Is the serial correlation in model or in model B? The period is 1949-1964 then n=16 years which
is the sample size in both the models. In model A we have only one explanatory variable K=1
given 5% significance level, n=16 &K=1 from the d-table dL=1-10 & du is 1.37. The calculated
d value of the regression model A is 0.8252 since it is less than the lower value of d-statistics
then we reject the null hypothesis & accept that in model A the random term Ui is positively auto
correlated with each others.
In case of model B we have two explanatory variables K=2, N=16 from the table at 5%
significance level dL=0.98 & du is 1.54. The calculated d-value of model B is 1.82 which is
greater than du then again here there is negative autocorrelation in random term Ui.

Haramaya University 116 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

5.2.3. Multicollinearity

In the assumptions of OLS we were assuming that the explanatory variables are not perfectly
linearly correlated. i.e.
Yi =  + X 1t +  2 X 2 t +  3 X 3t + Ut
Then we assume that the explanatory variables X1,X2 & X3 are not perfectly correlated means
rX1X2  1, r X1,X3  1, rX2,X3  0. But if it does not hold we speak of there is perfect
multicollinearity with the explanatory variables. If the explanatory variables are multicollinear
then we can not identify the independent effects of the explanatory variables on the explained
variables.
Suppose
Yt =  + 1 L +  2 k +  3 Z +  4T + Ut
Where L = lab our, K is capital, Z is land & T is technology Y is output.
Then if rLk=1 we can not trace the independent effects (contribution) of each in put factors on
output. When the explanatory variables are correlated with each other the method of OLS breaks
down. In theory we may have two situations in the relationship between explanatory variables.
rx1x2=1 i.e when the explanatory variables are perfectly correlated & the estimated parameters
using OLS method will be indeterminate i.e. ˆ will be indeterminate.
If rx1x20 if the explanatory variables are not inter-correlated with each other & this is called the
variables are orthogonal.
In practice neither of these extreme cases is often met. In most cases there is some degree of
inter-correlation among the explanatory variables due to interdependence of many economic
variables over time. The following basic points should be considered when we are dealing about
multicollinearity.
Multicollinearity is a sample phenomenon. When we assume a theoretical or population
regression model we believe that all explanatory variables included in the model have in
dependent effect (influences) on the dependent variable. But in this case we may have two
options.
In our sample all explanatory variables are so highly correlated & then we can not isolate their
individual influences on the dependent variable
In the sample some explanatory variables may be correlated with each other.
In any case we can not satisfy the assumption of independency among explanatory variables.
Multicollinearity is a question of degree & not kind. Means it is not a question of whether the
multicollinearity among explanatory variables is negative or positive. But what matters is the
extent the degree of correlation among the explanatory variables.
Multicollinearity is the problem arises due to the presence of linear relationship among the
explanatory variables.

5.2.3.1. Sources of multicollinearity

The problem of multicollinearity may arise for various reasons.

Haramaya University 117 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

1st Due to the inherent characteristics of economic variables they tend to move together overtime.
Means economic magnitudes are influenced by the same factors. Once such factors becomes
operative in the economy all variables tends to change in the same direction.
Ex. During boom period profit, income saving, output, employment, investment, consumption,
prices etc are tend to rises (move in the same direction) & decreases in period of recession. In the
time series data growth & trend factors are the main cause of multicollinearity.
2nd Multicollinearity may arise due to use of lagged explanatory variables in the regression
model.
Ex. Ct =  + 1Yt +  2Yt −1 + Ut
Where Ct=consumption expenditure.
Yt current income, Yt-1 is previous income. Here Yt & Yt-1 may be correlated with each others.
Hence the problem of multicollinearity may be observed in distributed lag models.
3rd Multicollinearity is usually connected with time series data because economic variables are
move together in the same directions.
4th – Multicollinearity is also quite frequent in cross sectional data where
Q= AL Keut
Where Qt= out put
Lt = lab our & Kt is capital
Then if you collect data from different manufacturing at a single period (at a time) then you will
find that in large firms capital & lab our are very high as comp aired to small firms i.e. capital &
lab our are tend to move in the same direction & will correlated with each other.
5th- An over determined model: - when the model has more explanatory variables than the
number of observations there will be multicollinearity. This could happen in medical research
where there may be a small number of patients about who information is collected on large
number of variables.

5.2.3.2. Consequence of multicollinearity

The presence of multicollinearity affects the least squares estimators & renders them inefficient.
The problem of multicollinearity must therefore be regarded as *a ‘block mark’ that reduces the
confidence in conventional test of least squares estimators.
The basic consequences of multicollinearity are
The least square estimators are indeterminate
The variance & covariance of the estimators becomes infinitely large means the standard errors
of these estimates will be very large.We prove there as follows
Suppose the relationship that will be estimated is
Yt =  + 1 X 1t +  2 X 2 t + Ut
Assume further that X1 & X2 are related with each other & their relation is X2=KX1 where K is
any arbitrary number. The formulae for the estimation of the coefficients ˆ1 & ˆ 2 are
( x y )( x ) − ( x y )( x x )
2

ˆ1 =
1 2 2 1 2

( x )( x ) − ( x x )
2
1
2
2 1 2
2

ˆ2 =
( x y )( x ) − ( x y )( x x )
2
2
1 1 1 2

( x )( x ) − ( x1x )
2
1
2
2 2
2

Haramaya University 118 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Now substitute KX1 in place of X2 & you will find that

ˆ1 =
( x y )( k x ) − ( kx y )( x kx )
1 1 1
2
1 1 1

( x )( kx ) − ( k x x )
2
1 1
2
1 1 2
2

k ( x y )( x ) − k ( x y )( x ) 0
2 2 2 2
1
= 1
=
1 1

k ( x ) − k ( x )
2 2 2
1
0 2 2 2
1

Again
( kx y )( x ) − ( x y )( x kx )
ˆ =  1
2
1 1 1 1

k ( x ) − k ( x )
2 2 2 2 2 2
1 1

( x y )( x ) − k ( x y )( x ) 2 2

=k = 0/0
1 1 1 1

k ( x ) − k ( x )
2 2 2 2 2 2
1 1

ˆ ,&ˆ 2 are indeterminate i.e. there is no way of finding separate values for ˆ ,&ˆ2 . The standard
error of the estimates becomes infinitely large.
Suppose we have
Yt =  + 1 X 1t +  2 X 2 t + Ut
If X1 & X2 are correlated (X2=KX1) then the variance of ˆ ,&ˆ2
ˆ
Will be Var ( 1 ) = u 2  x22
 x1  x22 − ( x1x2 )
2 2

Substitute in place of X2 = Kx1

ˆ
Var ( 1 ) = u 2  (kx) 22
 x1 ( kx1 ) − ( x1kx1 )
2 2 2

u 2k 2  x12
=
k 2  x1  x  x − k ( x )
2 2 2 2 2 2
1 1 1

  u
k2
= u 2 2
 k = x 2
2
2 2
1
=
k  ( x ) − ( x )  2 2
0 2 2
  1

Again to find the variance of Var ( ˆ2 )

Var ( ˆ2 ) =
x u 2 2
1

  x  x2 − ( x x2 )
2 2 2
1 1

Substitute KX1 in place of X2

ˆ u 2  x12
Var (  2 ) =
 x1 ( xk k ) −  x1 kx
2 2
( )
2

u 2  x12
=
k 2  x1
2
x x 2
1 1
2
− k2 ( x ) 1
2 2

Haramaya University 119 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

u 2  x 21
= =
0
Thus the variance of the estimates become infinite unless u 2 = 0
To summarize
Although we obtain OLS estimators, their standard error tends to be large as the degree of
correlation between the variables increase.
Because of large standard errors the probability of accepting a false hypothesis (i.e type II error)
increases.
The OLS estimates & their standard errors become very sensitive to slightest change in the
sample data.
If multicollinearity is high, one may obtain a high R2 but none of them or very few estimated
regression coefficients are statistically significant.
5.3.3 Test for detecting multicollinearity
1) Method based on Frisch's confluence analysis or Bunch - map analysis. According to this test
the seriousness of multicollinearity may be detected by
coefficient of determination R2
Partial correlation coefficient  x1x2
Standard error of the regression coefficients.
But none of these symptoms by itself is a satisfactory indicator of multicollinearity because
large standard errors may arises for various reasons not only due to the presence of
multicollinearity
high  x1x2 is not only a sufficient but no a necessary condition for the existence of
multicollinearity
R2 may be high & yet the estimates may not be significant & imprecise.
However a combination of this entire criterion should help the detection of multicollinearity, as
has been suggested in Frisch's test. In order to gain as much knowledge as possible as to the
seriousness of multicollinearity it is suggested the following method.
If the regression model is assumed to be
Yi =  + X 1 + X 2t + X 3 + Ui
Means that Y= f(X1X2X3) then the procedure is regress Y on X1, X2, &X3 separately i.e. stepwise
first regress Y on X1 then on X2 & finally on X3. Thus
In elementary regression we examine their results on the basis of a priori & statistical criterion.
We choose the elementary regression which aspects to give the most possible results on both a
priori & statistical criterion. Then gradually insert additional variables & we examine their
effects on the individual coefficients i.e. standard error & R2. And the new variable is classified
as useful, superfluous or detrimental as follows.
If the new variable improves R2 without rendering the individual coefficients unacceptable
(wrong) on a priori considerations, the variable is considered useful & is retained as an
explanatory variable.
If the new variable improves R2 and does not affect to any considerable extent the value of the
individual coefficients, it is considered as superfluous & is rejected.
If the new variable affects considerably the sign or values of the coefficients it is considered as
detrimental.
If the individual coefficients are affected in such away as to become unacceptable on theoretical,
a priori, considerations, then we may say that this is a warning for the existence of

Haramaya University 120 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

multicollinearity. In this case the new variable is important, but because of intecorrelation with
the other explanatory variables its influence can not be assessed statistically by OLS. This does
not mean that we must reject the detrimental variable.If we omit the detrimental variable
completely to avoid multicollinearity.
we must bear in mind that the influence of the detrimental variable will be absorbed by other
coefficient and
The influence may be absorbed by the random term which may become correlated with the
variable left i.e. E(XUi)  0. It will violate the assumption of OLS which is Cov (UiXi)=0
2) Examination of partial correlations. If in the regression Y on X2,X3 & X4 the R2 is very high
but 2X3X4, 2X2X4, 2X2X3, are comparatively low, this may suggest that the variables X2,X3 &
X4 are highly interconnected.
Auxiliary regression. Here we regress each X on the remaining X variables & compute the
corresponding R2, which we denote by R2i. Each one of these regression is called auxiliary
regression. That is auxiliary to the main regression of Y on the X variables. Then based on the
relation between the F test & R2.

R 2 x1 x 2 , x3 ...xk
Ri = k −2
1 − R 2 x1 , x 2 , x3 ...xk
(n − k + 1)
Follow the F distribution with k-2-numerator
N-k+1 denominator degree of freedom. In the equation
N-stands number of sample size
k- Number of explanatory variables including the intercept term
R2 X1X2...Xk the coefficient of determination in the regression of variable Xi on the remaining X
variables. If the computed F at chosen level of significance exceeds the table value of F it
would indicate the presence of multicollinearity between Xi & the other X variables. If it does
not, we say that the particular Xi is not collinear & we may retain the variable in the model. This
rule is some how simplified if we use Klein’s rule of the thumb which states that we except
multicollinearity if the R2 computed from the auxiliary regression is greater than the over all R2
obtained from the regression of Y on all repressors. The above test of multicollinearity will show
us the location of multicollinearity.
Tolerance & Variance inflation factor. The variance & covariance of the multiple regression i.e.
Yi =  + 1 X 1i +  2 X 2 + Ui
Can be written as
u 2
Var ( ˆ1 ) =
 x12 i(1 − r 2 x1 x2 )
u 2
Var ( ˆ2 ) =
 x i(1 − r
2
1
2
x1 x2 )
− rx1 x2u 2
Cov( ˆ2 ˆ1 ) =
(1 − r 2 x1 x2 )  x i. x i
2
1
2
2

From the above if x1x2 tend to wards 1 that is, co linearity increases, the variance of the two
estimator increases & in the limit x1x2=1 the variance of the estimators will be infinite. It is

Haramaya University 121 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

equally clear from the above formulae as x1x2 increases towards 1, the covariance of the two
estimators also increase in absolute value. The speed with which the variance & covariance
increases can be seen with the variance inflation factor (VIF) which is defined as
1
VIF =
(1 − r 2 x1 x 2 )
VIF shows how the variance of an estimator is inflated by the presence of multicollinearity. As
2x1x2 approaches to 1. The VIF approaches infinity. As the extent of co linearity increases, the
variance the estimator increases, & in the limit it can become infinite. If there is no co linearity
x1x2=0 then VIF will be 1. Then we can write the variance of follows
2
Var ( ˆ1 ) = VIF
 x12i
2
Var ( ˆ 2 ) = VIF
 x22 i
This shows the variance of ˆ ,&ˆ 2 are directly proportional to the VIF. We could also use what
is known as the measure of tolerance which is defined as
TOLJ = (1 − R 2j ) = 1
VIF j
From which TOLj is 1 if there is no co linearity where as it is zero when there is perfect col
linearity.
Computation of t- ratio to the pattern of multicollinearity. The t- test helps to detect those
variables which are the cause of multicollinearity. The test is performed on the partial correlation
coefficients through the following procedure of hypothesis testing.
The null hypothesis is H0= xixj.x1x2x3...xk=0
The alternative is H1= xixj.x1x2x3...xk0
In three variable models
(r12 − r13 .r23 ) 2
2x1,x2x3=
(1 − r13 )(1 − r23 )
2 2

(r13 − r12 .r32 ) 2

2x1,x3.x2=
(1 − r12 )(1 − r32 )
2 2

(r23 − r21.r31 )2
 x2,x3. x1=
2
(1 − r21 )(1 − r31 )
2 2

The form this we can compute the t-test for each estimator as follows
(r 2 xi x j .x1 , x2 , x3 ...xk ) n − k
t* =
1 − r 2 xi x j .x1 , x2 ,...xk
If calculated t* is > the table value of t reject the null hypothesis that no-multicollinearity &
accept that there is multicollinearity. If calculated t* t table value accept the null hypothesis that
there is no multicollinearity & reject he alternative that there is multicollinearity.

Haramaya University 122 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

5.2.3.3. Solutions to the problem of multicollinearity

The solutions which may be adopted if multicollinearity exists in a function vary depending up
on
severity of the multicollinearity
availability of other sources data
on the importance of factors which are multicollinear
On the purpose of which the function is being estimated & other conditions.
To solve these problems
Some writers have suggested that if multicollinearity does not seriously affect the estimates of
the coefficients, one may tolerates its presence in the function, although the integrity of OLS
estimates is to a certain extent impaired.
Others suggested that if multicollinearity affects some of the unimportant factors one may
exclude these factors from the function. Again specification error may be expected to undermine
the BLUE character of the ordinary least squares.
Multicollinearity may affect only a part of ˆ 's while other estimates may remain fairly stable &
reliable. In this case
The reliable ˆ 's may be used for any purpose. i.e forecast or policy formulation
All the estimates may be used for forecasting provided that the same multicollinearity pattern
will continue to exist in the forecast period.
Given all the above if multicollinearity has serious effect on the coefficients estimates of
important variables one should adopt any one of the correlation solution.
Ridge Regression: - In this case just added a constant number  to the variance of the
explanatory variables before solving the normal equations. This is found to decrease the problem
of multicollinearity but it is a mechanical solution as there is no clear cut method of solving.
Increase the size of the sample!- To avoid multicollinearity it is suggested that multicollinearity
may be avoided or reduced by increasing the sample size. The reason for this is that as you
increase the sample size the covariance among the parameters (X's) get reduced. But one should
remember that this should be true when inter correlation happens to exist only in the sample but
not in the population, the procedure of increasing the size of the sample will not help to reduce
the multicollinearity.
Dropping a variable (s):- The problem of multicollinearity may be reduced or solved if we drop
the variable(s) that is (are) collinear. But here we must be carefully not to commit a specification
bias or specification error.
Let Yi = 1 X 1 +  2 X 2 + Ui
From this model
E ( ˆ1 ) = 
 2u
Var ( ˆ 1) =
 xi 2
(1 − r 2 x1x 2 )
Now let’s omit variable X2 by expecting that it is collinear with X1 & we will have the following
model
Yi = 1 X 1 + +Ui
Model after omitting X2, the estimator will have

Haramaya University 123 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

ˆ1 * =
X Y 1 i

X
2
1

In place of Yi substitute 1 X 1 + ̂ 2 X 2 + U
ˆ1  X 1 X 1 + ˆ2  X 1 X 2 +  X 1Ui
ˆ1 * =
 X1
2

We know that from our assumption  X Ui =0 then

1

ˆ1 * = ˆ1
X 1
2

+ ˆ 2
X X 1 2

X X
2 2
1 1

E ( ˆ1 *) = ˆ1 + ˆ 2
X X 1 2

X
2
1

 X 1u  u 2  X 1 2 u 2
ˆ
Var ( 1 *) = Var   = =
  Xi 2 
   X1
2 2
 X2
2
( )
Thus the estimator after omitting the variable X2 is biased (large mean) but has smaller variance
than ˆ1 (completed model variance). Therefore, dropping a variable with the hope to estimate the
problem of multicollinearity may lead to biasness of estimators.
Introducing an additional equation in the model. The problem of multicollinearity may be
overcome by expressing the relationship between multicollinear variable. Such relation in a form
of an equation may then be added to original model. The addition of null equation transforms our
single model (original) in to simultaneous equation model. The reduced form method can then be
applied to avoid multcollinearity.
Use extraneous information:- Extraneous information means the information obtained from any
other source outside of the sample which being used for estimation. Extraneous information may
be available from economic theory or from some empirical studies already conducted in the field
in which we are interested. We can use the following methods which extraneous information is
utilized in order to deal with the problem of multicollineaity.
A priori information. Suppose we consider the following model
Yi =  + 1 X 1i +  2 X 2 + Ui
Where Yi= consumption X1i=income & X2i wealth. Income & wealth variables may move
together & create multicollinearity. But suppose in a priori if we know that  2 = 0.101 i.e the
th
1
rate of change of consumption with respect to wealth is of the corresponding rate with
10
respect to income. Then we can write the regression
Yi =  + 1 X 1i + 0.101 X 2 i + Ui
Yi =  + 1 ( X 1i + 0.10 X 2 i) + Ui
Let X1i + 0.10X2 = X*1 substitute
Yi =  + 1 X 1 * +Ui
Run regression on the model /apply OLS/ & obtain the estimator 1 & from the relationship
between ˆ1,&ˆ . You can calculate ̂ from ̂ 2 =0.10 1
2 2

Haramaya University 124 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Combining cross sectional & time serious data or pooling cross section & time series data.
Suppose we want to study the demand function for automobile & assume that we have a time
series & cross sectional data; on the number of cars sold, average price of the cars, & consumers
income. Suppose we have the following
Yt = Pt  1 It  2 et
ut

Given that the demand function is studied using the above model. Where. Yt= demand for cars
(number of cars sold), Pt is average price of the cars, I is income & t is time. If our objective is to
estimate the price elasticity of demand (i.e. 1 ) & income elasticity using  2 . First transform the
non-linear function in to linear function i.e.
ln Yt = ln  + 1 ln Pt +  2 ln It + ut.
In time series data the price & income variables generally tend to be highly collinear i.e. one can
not separate the income & price effect on quantity demanded. On the other hand it is not possible
to obtain price effect B1 from the cross sectional data (because price structure is the same for all
consumers at a particular point of time). Under such condition it is suggested to use pooling
techniques which avoids to a certain extent the problems associated with both (cross & time
series) of sample data. Pooling techniques can be outlined as follows.
1st Cross-section sample is used to obtain an estimate of the income coefficient using the
following
ln Yt =  +  2 * ln It + Ut
And find the influence of change in income (It) on demand Yt is eliminated.
Zt =Yt -- ̂ 2 It
2nd stage. The new variable obtained (Z) is regressed on price variable using time series data to
estimate price coefficients ̂ 2
Z t =  +  , ln Pt + Ut
By combining the result, the relationship becomes
ln Yt = ln ˆ + ˆ1 ln Pt + ˆ 2 * It
Where ˆ is derived from the time series data & ˆ * is obtained from the cross sectional data.
1 2
By pooling techniques we have skirted the multicolinearity between income & price. The
estimators obtained using pooling the time - series and cross-sectional data in the manner just
suggested may create problems of interpretation. Because we are assuming implicitly that the
( )
cross sectional estimated income elasticity ˆ2 * is the same thing as that which would be
obtained from a pure time series analysis.
Transformation of variables: Given that we have time series data on consumption (Ct), income
(UE) & wealth (W).
Ct =  + 1Yt +  2Wt + Ut
From the above model income & wealth may tend to move in the same direction & creates
multicollinearity. One way of minimizing this dependence is to proceed as follows. Take the
lagged (t-1) values of the above model & you will have
Ct − 1 =  +  2Yt − 1 +  2Wt −1 + Ut − 1
Subtract from the previous model
Ct − Ct − 1 = ( −  ) + 1 (Yt − Yt − 1) +  2 (Wt − Wt − 1) + (Ut − Ut − 1)
Let C*t = Ct-Ct-1

Haramaya University 125 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Y*t = Yt-Yt-1
W*t= Wt-Wt-1
Vt= Ut-Ut-1
Ct * = 1Y * t +  2Wt * +Vt
This equation is known as the first difference form because we run the regression, not on the
original variable, but on the difference of successive values of the variable. The first difference
regression model often reduces the severity of multicuollinearity because there is no a priori
reason to believe that the first difference model will also be highly correlated. But the problem
with this difference transformation model is the error terms Vi may not satisfy one of the
assumption of CLR i.e. the disturbance term are not serially correlated.

Haramaya University 126 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

5.3. Exercise for chapter

1) a) What is meant by perfect multicolinarity? What is its effect?
b) What is medant by high multicollinarity but not perfect multicollinearity?
c) How can multicollinearity be detected? How can you avoid or reduce ulticollinearity?
2) Aresearcher collects the following data from 15 firms about their capital, labour & out put.

Firms Output in tones Labour employeed Capital

1 2350 2334 1570
2 2470 2425 1850
3 2110 2230 1150
4 2560 2463 1940
5 2650 2565 2450
6 2240 2278 1340
7 2430 2380 1700
8 2530 2437 1860
9 2550 2446 1880
10 2450 2403 1790
11 2290 2301 1480
12 2160 2253 1240
13 2400 2367 1660
14 2490 2430 1850
15 2590 2470 2000

Fit a cobbdoglas production function find te coffeicent of determination?

Regress Q on L only
Regress Q on K
What can you say about the multicollinearity?
2) a) What is meant hetroscedasticity?
b) Why hetroscedasticity is a problem?
c) How the presence of hetroscedasticity is tested?
d) Given the following data on consumption expenditure C & disposable income Yd for 30
families
Family Consumptio Yd Family Consumption Yd
1 10600 12000 16 14400 17000
2 10800 12000 17 14900 17000
3 11100 12000 18 15300 17000
4 11400 13000 19 15000 18000
5 11700 13000 20 15700 18000
6 12100 13000 21 16400 18000
7 12300 14000 22 15900 19000
8 12600 14000 23 16500 19000
9 13200 14000 24 16900 19000
10 13000 15000 25 16900 20000

Haramaya University 127 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

11 13300 15000 26 17500 20000

12 13600 15000 27 18100 20000
13 13800 16000 28 17200 21000
14 14000 16000 29 17800 21000
15 14200 16000 30 18500 21000
Regress consumption on disposable income & test the exisitance of hetroscedisticity?
If there is hetroscedasticity correct it using appropriate method?( assuming the variance is
proportional to Yd2( square of the disposable income)a) What is meant autocorrelation?

a) Why autocorrelation is a problem?

b) How can autocorrelation be corrected?
c) How is the presence of positoive or negative first order autocorrelation tested?
d) The presence of autocorrelation can be tested by calculating the Durbin-Watson statistic.
e) Given the following data on import & GDP from 1980 -1999. Then regress M on GDP & test
for autocorrelation at 5% level of significance?

Year Imports GDP

1980 299.2 2918.8
1981 319.4 3203.1
1982 294.9 3315.6
1983 358 3688.8
1984 416.4 4033.5
1985 438.9 4319
1986 467.7 4537.5
1987 536.7 4891.6
1988 573.5 5258.3
1989 599.6 5588
1990 649.2 5847.3
1991 639 6080.7
1992 687.1 6469.8
1993 744.9 6795.5
1994 859.6 7217.7
1995 909.3 7529.3
1996 992.8 7981.4
1997 1087 8478.6
1998 1147.3 8974.9
1999 1330.1 9559.7

f) Correct autocorrelation if it is found in the above data?

Haramaya University 128 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Chapter 6 : Estimation with Dummy Variables

6.1. Introduction
The variables used in regression equation usually take values over some continuous range. In
regression analysis the dependent variable is frequently influenced not only by variables that are
quantitative (income, output, prices, costs, heights etc.) but also by variables that are essentially
qualitative in nature (sex, race, profession etc.). Dummy variables are constructed by
econometricians to be mainly used as proxies for other variables which cannot be measured in
any particular case for various reasons. Dummy variables are commonly used as proxies for
qualitative factors such as sex, profession, religion etc. Since such qualitative variables usually
indicates the presence or absence of a quality or an attribute such as male or female, black or
white, etc. One method of quantifying such attribute is by constructing artificial variables that
take on values of 1 or 0. 1 indicates the presence of an attribute & 0 indicates the absence of an
attribute. Variables that assume 1 or zero values are called dummy variables, or indicator
variables, binary variables and dichotomous variables.

Suppose the firm utilized two types of production process to obtain its output.

Yi =  + 1 X 1t + Ut
Where Yi is output obtained
X1 is a dummy variable.
1 if the output is obtained from machine A
Xi= 0 if the output is obtained from machine B

By applying OLS you obtained ˆ & ˆ .

• ̂ Measures the mean excepted output associated with machine B
• ˆ Measures the difference in mean output associated with a change from machine B to
machine A. (by how much the mean output of machine B is deviating from the mean output
of machine A).
• ˆ + ˆ the mean expected output from machine A.

The dummy variable procedure can easily be modified if more than two distinct values are
involved. Assume that in the above example three different processes may be employed & one
wishes to account that the three processes may not give identical output. Then in this case we
constructed two dummy variables to take in to account the three different process of production
(Always when there are N variables we develop N-1 dummy variable.)

Suppose
Yt =  + 1 X 1t +  2 X 2 + Ui
The output is produced using three methods i.e A, B & C.
➢ X1t= 1 if the output is obtained from machine A and 0 otherwise (if it comes from B & C the
value will be zero).

Haramaya University 129 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

➢ X2t = 1 if the output is obtained from machine B & otherwise zero. ( if it comes from A&C
the value will be zero).
➢ In this case ̂ represents the mean value of output obtained from machine C.
ˆ1 = represents the difference in output associated with a change from machine C to
machine A.
ˆ + ˆ = the mean value output obtained from machine A.
ˆ 2 represent the difference in output associated with a change from machine C to
machine B.
ˆ2 + ˆ = the mean value of output obtained from machine B.

The above examples consist of dummy variables as explanatory variable. Such models are called
analysis of variance (ANOVA) i.e the dependent variable is quantitative but the explanatory
variables are qualitative. But in most economic research, regression model contains a mixture of
quantitative and qualitative variables. Such as

Ct =  + 1 x1t +  2 Dt + Ut
Where: Ct = consumption expenditure
X1t = income of the consumer
Dt= is a dummy variable indicating windfall gain and loss.
In this case we have quantitative and qualitative variables as explanatory variables. Such models
are called Analysis of covariance (ACOVA).

Ct =  + 1Y1t +  2 D1t +  3 D2t +  4 D3t + Ut

Where Ct= consumption expenditure of the house hold
1 if the house hold has children
D1t=
0 if no children (otherwise)

1 if the house hold owns its own house

D2t=
0 (otherwise)

1 if the age of the house hold is over 70 years

D3=
0 (otherwise)

Notice that each of the three qualitative variables, having children, owning house, and age of the
house hold, has two categories and hence need one dummy variable for each. Not also that the
omitted or base, category now as “a house hold with no child, no house and less than 70 years” is
given by the following equation.

Ci =  + 1Yt + Ut
To find the mean consumption of this house hold is given by

E(Ct)=  + Yt + Ut
E(Ut)=0

Haramaya University 130 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

E(Ct) =  + Xt
Mean value of the consumer who has a child but no house & less than 70 years. i.e. D1t=1, D2t=0
& D3t=0. Then
Ct =  + 1Yt +  2 D1t + Ut
E (Ct ) = E + 1Yt +  2 D1t + Ut 
E (Ct ) =  + 1Yt +  2 D1
Because E (Ut)=0

Again this is the mean consumption expenditure of the consumer. By the same analogy you can
continue in such away.

6.2. Some important uses of Dummy variables

Analysis of Covariance model
Use of dummy variable for measuring the shift of a function. A shift of a function implies that
the constant intercept changes in a different period while the other coefficients remaining the
same. Such type of shift can be examined by introducing a dummy variable in the function under
study. Suppose if we want to study the consumption expenditure of the society as a whole for the
period 1910-1950. We know that there were two break through events where occurring during
these periods. These are.
1st world war in 1910’s & 2nd WW in 1940’s
Great depression was occurring in 1930’s
Given these two events the remaining period is assumed to be normal years. Then generally we
can divided these period as normal years & abnormal (war & depression period) years. During
these periods it is expected to have different consumption expenditure patterns of the society.
Suppose we assume that during the abnormal & the normal period marginal propensity to
consume is constant but there is a change with respect to the autonomous /subsistence level/
consumption. The consumption function will be

Ct =  + 1Yt +  2 Dt + Ut
Where Ct = consumption expenditure
Yt= income of the consumers
1 for normal years
Dt=
0 for abnormal years

The normal years function (when Dt=1) all periods)

Ct =  + 1Yt +  2 + Ut
Ct = ( +  2 ) + 1Yt + Ut
Given that  +  =  then
Ct =  + 1Yt + Ut
Cˆ t =  Yt
1
The abnormal years function (when Dt=0) would be
Ct =  + 1Yt +  2 + Ut
Because Dt=0 in all periods

Haramaya University 131 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Cˆ t =  + 1Yt
Then we apply OLS using two explanatory variables Yt & Dt & obtain estimates of Yt & Dt in
the regression equation i.e.

Ct =  + 1Yt +  2 Dt + Ut
Apply OLS (Suppose you get the following equation)

Cˆ t = 19.5 + 0.85Yt + 10Dt

Now from this estimated you can get the normal & abnormal year period equations as follows. If
there is war period /abnormal year/ Dt=0 & the equation will be
Cˆ t = 19.5 + 0.85Yt Abnormal year equation.
If there is normal year Dt will have a value of 1 then
Cˆ t = 19.5 + 0.85Yt + 10Dt
Dt= 1 in all periods
Cˆ t = 19.5 + 0.85Yt + 10
Cˆ t = (19.5 + 10) + 0.85Yt
Cˆ t = 29.5 + 10Yt
To generalize
Cˆ t = 19.5 + 0.85Yt Abnormal year
Cˆ t = 29.5 + 0.85Yt Normal year
These function suggest that there is a difference in the intercept in abnormal & normal period
since the abnormal period is less than the normal period intercept then we can conclude the
depression & the two war periods had significantly negative effect on consumption expenditure

Y
Cˆ t = (ˆ0 + ˆ2 ) + ˆ1Yt Normal
Ĉ Cˆ t = ˆ + ˆ1Yt Abnormal year

ˆ + ˆ2
0 income X

ˆ

B) Use of Dummy variable for measuring the change in the slope parameter over time. The
abnormal period may not affects the autonomous (subsistence level) but it may affect the
marginal propensity to consume i.e. if the abnormal periods affects the marginal propensity to
consume but not the constant term. Suppose we have
Haramaya University 132 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Ct =  + 1Yt +  2 (YtDt) + Ut

1 in normal years
Dt=
0 in abnormal years

Since we assume that the abnormal year is affecting the slope (MPC) we clipped it with
disposable income. If the period is normal year the estimated function will be (Dt=1)

Ct = ˆ + ˆ1Yt + ˆ2Ytˆ
Cˆ t =  + ( 1 +  2 )Yt Normal year equation. If the period is abnormal year i.e Dt=0 then the
estimated function will be

Cˆ t = ˆ + ˆ1Yt Abnormal year equation.

To have both the abnormal & normal year estimated function first we use data to estimate

Ct =  + 1Yt +  2 ( ytDt) + Ut
Cˆ t = ˆ + ˆ1Yt + ˆ 2 (YtDt) Will be the estimated function.
On the basis of statistical criterion check all the parameters (ˆ , ˆ1 ˆ 2 ) significance. Then when
Dt= 1 you will have

Cˆ t = ˆ + ( ˆ1 + ˆ 2 )YtDt
let ˆ + ˆ = ˆ *
1 2

Cˆ t =  + ˆ1 * YtDt This is the normal year equation.

When Dt=0 the estimated function will have
Cˆ t = ˆ + ˆ1Yt This is for abnormal year.
Since ˆ* = ( ˆ1 + ˆ2 ) is greater than ˆ1 we can conclude that even though the constant term
(autonomous consumption) in both periods (ˆ ) is equal but the marginal propensity to consume
is different in normal & abnormal period

Haramaya University 133 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Ĉ t
Normal year Cˆ t = ˆ + ˆ * YtDt

Abnormal year Cˆ t = ˆ + ˆ1Yt

̂ Income

C) The final possibility is when there is a change in the intercept & slope over time. The
regression equation when it is considering two things simultaneously i.e affecting the slope
coefficient & the constant term can be explained using the following function.
Ct =  + 1Yt +  2 Dt +  3Yt Dt + Ut
Dt=1 for normal year
Dt=0 for abnormal year.
Then the function would be
Normal year when Dt=1
Ct =  + 1Yt +  2 (1) +  3Yt(1) + Ut
Ct =  + 1Yt +  2 +  3Yt + Ut
Ct = ( +  2 ) + (1 +  3 )Yt + Ut
let  1 =  + 2 &  2 = 1 + 3
Ct =  1 +  2Yt + Ut
Cˆ t = ˆ + ˆ Yt
1 2
Abnormal year when Dt=0
Ct =  + 1Yt +  2 (0) +  3Yt(0) + Ut
Ct =  + 1Yt + ut
Ct = ˆ + ˆ Yt
1

Haramaya University 134 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

Normal year C t = ˆ + ˆ1Yt

Abnormal year Cˆ t = ˆ + ˆ * YtDt

( ˆ + ˆ1 )

 Income

6.3. Use of Dummy variables in Seasonal analysis

To know how the dummy variables are used to capture the seasonal effects (quarters) in the
economic time series data we can form the regression equation as follows. We have four quarters
in a year these are quarter one, two, three & four.
To test seasonal pattern using intercept

Yt = i + 1 X 1t +  2 D1 +  3 D2t +  4 D3t + Ut
Where Yt= is volume of profit
Xt = is Volume of sales

1 if it is the second quarter

D1=
0 otherwise

1 if it is the third quarter

D2= 0 otherwise

D3= 1 if it is the fourth quarter

0 otherwise

If there are seasonal patterns in various quarters the estimated differential intercept will explain it
by applying OLS you can estimate the function & obtain the coefficients.
Yˆt = ˆ + ˆ1 xt + ˆ2 D2 + ˆ3 D2 t + ˆ4 Dt
If D1= 1 we could find that (second quarter)
Yˆt = ˆ + ˆ1 xt + ˆ2 D1

Haramaya University 135 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

= (ˆ + ˆ 2 ) + ˆ1 Xt
If D2 =1 (third seasons)
Yˆt = ˆ + ˆ1 xt + ˆ3 D2
= (ˆ + ˆ ) + ˆ Xt
3 1
If D4 =1 (fourth quarter)
Yˆt = ˆ + ˆ1 Xt + ˆ4 D3
Yˆt = (ˆ + ˆ ) + ˆ X
4 1 1
If it is in the first quarter
Yˆt = ˆ + ˆ Xt

Because all the other quarters will be assigned zero value. In the above case we consider only the
intercept term which indicates the presence of seasonal patterns
Ex. If we obtain the following results from a given data.
Yˆt = 6688+ 1322D2- 217D3t+ 183D4 0.0383Xt.
S.E (1711) (638) (632) (654) (0.0115)
t (3.908) (2.07) (0.344) (0.281) (3.33)
R2=0.5255

The results shown that only the sales coefficient (Xt) & the differential intercept associated with
the second quarter are statistically significant at 5% level. It means on the basis of S.E & t-test
you can check it as follows
ˆ
S .E (ˆ )  ,
2
,
ˆ ˆ1
S .E ( 1 ) 
2
ˆ
S .E ( ˆ 2 )  2 ,
2
ˆ
S .E ( ˆ1 ) 
2
ˆˆ

S .E ( ˆ 4 )  4
2

And in all case if t>2 you can accept that the explanatory variables will affect the dependent
variables (i.e. Profit is affected by sales (Xt), & the dummy variables). In many of the study only
some of ̂ ‘s are significant which reflects that only some quarters may affect the profit From the
above example only ̂ , ̂ 3 are statistically significant. Thus we can conclude that there are some
seasonal factors operating in the second quarter of each year but there is no seasonal factors that
will affect profit in the other quarters. The average profit in the first quarter is 6688 & in the

Haramaya University 136 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

second quarter the average profit is ( ̂ , + ˆ3 ) i.e 6688+1322 = 8010. The sales (X1) coefficient
ˆ tells us that if sales increase by 1 average profits will increase by 0.033 cents.
1
If the seasonal factor do not affect the mean value but the slope of the seasonal factors.
Yt = t + 1 X t +  2 D1 Xt +  3 D2 Xt +  4 D3t + Ut
Where
1 If it is the second quarter
D1= 0 Otherwise

1 If it is the third quarter

D2= 0 Otherwise

1 If it is the fourth quarter

D3= 0 Otherwise

The seasonal effect can be captured by the slope coefficients. If all (D1, D2, D3 are zero)
Yˆt =  +  Xt + Ut
This is the first quarter (the base quarter.)
If D1=1 (second quarter)
Yt =  + 1 Xt +  2 D1 Xt + Ut
Yˆt =  +  xt +  D xt + ut
1 2 1

Yt =  + ( 1 +  2 ) Xt + Ut
If D2 = the third quarter
Yt =  + 1 Xt +  3 (1) Xt + Ut
Yt =  + (1 +  3 ) Xt + Ut
If D3 = 1 the forth generator
Yt =  + ( 1 +  4 ) Xt + Ut
Now the seasonal effect can now be examined by using hypothesis testing of F-test to know the
joint values are equal to zero 2 = 3 = 4 = 0
The final possibility is that let the intercept and the slope will be affected by the seasonal factors.
And the function would be
Yt =  + 1 Xt +  2 D1 Xt +  3 D2 Xt +  4 D3 Xt +  5 D1 +  6 D2 +  7 D3 + Ut
Given the previous meanings for all variables
Yt =  + 1 Xt is for the base quarter (1st quarter)
If D1= 1 for the second quarter
Yt =  + 1 Xt +  2 D1 Xt +  5 D1 + Ut
= ( +  5 ) + (1 +  2 ) Xt + Ut
If D2= 1 for the third quarter
Yt =  + 1 Xt +  3 D2 Xt +  6 D2 + Ut
= ( +  6 ) + (1 +  3 ) Xt + Ut

Haramaya University 137 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

If D3= 1 for the fourth quarter

Yt =  + 1 Xt +  4 D3 Xt +  7 D3 + Ut
= ( +  7 ) + (1 +  4 ) Xt + Ut

6.4. Exercise for the chapter

1) The following table gives the consumption expenditure C, the disposable income Yd & sex
of the house hold head of 12 random families.
Family Consumption Disposable income Sex type
1 18535 22550 M
2 11350 14035 M
3 12130 13040 F
4 15210 17500 M
5 8680 9430 F
6 16760 20635 M
7 13480 16470 M
8 9680 10720 F
9 17840 22350 M
10 11180 12200 F
11 14320 16810 F
12 19860 23000 M

a) Regress Yd on X
b) Test for a different intercept for families with a male or a female as head of the household.
c) Test for different slope or MPC for families with male or female as head of the household.
d) Test for both different intercept & slope.
e) which is the best result

Haramaya University 138 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

2) Given the following sales in four quarters from 1995 -1999

Year Quarter Time trend Sales
2010 1 1 540.5
2010 2 2 608.5
2010 3 3 606.6
2010 4 4 648.3
2011 1 5 568.4
2011 2 6 632.8
2011 3 7 626
2011 4 8 674.6
2012 1 9 587
2012 2 10 640.2
2012 3 11 645.9
2012 4 12 686.9
2013 1 13 597
2013 2 14 675.3
2013 3 15 663.6
2013 4 16 723.3
2014 1 17 639.5
2014 2 18 716.5
2014 3 19 721.9
2014 4 20 779.9

Using the above data run regression of sales and the seasonal dummies & interoperate the result?

Haramaya University 139 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])

 lOMoARcPSD|43275824

Introduction to Econometrics for Management

REFERENCE

✓ Griffiths, W. Using EViews: For principles of econometrics, 4th ed., 2011

✓ Gujarati, D., Basic econometrics, 5th ed. 2008.
✓ Gujarati, D., Econometrics by example, 2011.
✓ J. H. Stock and M. W. Watson, Introduction to Econometrics, 3rd revised edition
✓ Jeffrey M. Wooldridge Introductory Econometrics: A Modern Approach, 6th edition
✓ Jeffrey Wooldridge 2009 or 2012, Introductory Econometrics, 4th or 5th edition, Thomson
South Western
✓ Kennedy, P., 2008, A Guide to Econometrics, 6th edition, Wiley.
✓ Maddala, G. (2009), Introduction to Econometrics, 4th Edition, Wiley.
✓ Studenmund, A. (2011), Using Econometrics: A Practical Guide, 6th Edition, Pearson.
✓ Stundenmund, A.H., Using econometrics: A practical guide, 5th ed. 2005. Wooldridge, J.,
Introduction to econometrics: A modern approach, 4th ed. 2008.
✓ Thomas, R.L. (1996), Modern Econometrics: An Introduction, Addison-Wesley

Haramaya University 140 Distance Education Program

Downloaded by Biruk Bizuneh ([email protected])