ECONOMETRICS LECTURE NOTES

Definition: is the application of statistical and mathematical tools in the analysis of

economic data in order to;

a) Add empirical content to economic theory

b) Validate or refute the theories.

Econometrics can also be defined as the art and science of using economic theory and
statistical techniques to analyse economic data.

Discipline Interrelationship

Overview

- Economic Theory (Economic Model) – gives relationship among economic

variables, the relationship is captured in an economic model and does not talk
about the measurability of the variables. Economic models are based on
abstractions from reality with those variables not included in the model assumed
to be important.
- Transform the economic model into a mathematical model – mathematical
models yield deterministic models and are not measurable using empirical data.
Mathematical models state an exact relationship of the variables.
- An econometric model is a mathematical model plus residuals and residuals
capture all unobservable variables that impact the dependant variable. An
econometric model is given as:

𝑃𝑡 = 𝛼0 + 𝛼1 𝑀𝑡 + 𝜇𝑡

- Economic Statistics – collects, analyses and presents data on the economic

variables in the form of a table and charts.
- Mathematical statistics – gives tools used to estimate/present economic data.
- Econometricians use observed data and not experimented data.

Covariance

𝛿𝑋𝑌 = 𝐸 (𝑋 − 𝜇𝑋 )(𝑌 − 𝜇𝑌 )

Problem: covariance is sensitive to the units of measurement used.

In economics we are interested in finding out the causal relationship of variables – no

aspect of causality is implied by the covariance measure.

Correlation

Rectifies the problem of covariance.

Page 1 of 16
 𝛿𝑋𝑌
𝑟𝑋𝑌 =
𝛿𝑋 𝛿𝑌

This is a scaled covariance of the two variables. Correlation measure is scale free since
the numerator and the denominator are measured in scales. The correlation measure
lies between minus one and one. −1 < 𝑟𝑋𝑌 < 1

0 = No relationship.

Direction for causation not captured in the covariance is captured in the correlation.

Regression Measure

Used to measure the degree of the direction of causality.

Types of Econometrics

Econometrics

theoretical Empirical

classical Bayesian classical Bayesian

Theoretical economics – concerned with identifying appropriate techniques for the

analysis of economic theories e.g. least squares estimation techniques, ordinary least
squares summarised into generalized least squares.

Theoretical economics must then spell out the assumptions of the techniques, the
properties, and what happens when one of the assumptions is violated.

Empirical Economics – concerned with using the tools developed in theoretical

econometrics into appropriate areas of study (applied econometrics)

Steps in Regression Analysis

1. Statement of economic theory/hypothesis

𝑃 = 𝑓(𝑀)

Page 2 of 16
 Too much money chasing too few goods is the cause of inflation
2. Specification of a mathematical model for the theory
𝑃𝑡 = 𝛼0 + 𝛼1 𝑀𝑡
The nature of the model is based on the assumption and usually does not specify
the nature of the equation
3. Specification of the econometric model; 𝑃𝑡 = 𝛼0 + 𝛼1 𝑀𝑡 + 𝜇𝑡
𝜇𝑡 = error term, residual, stochastic term.
4. Data Collection
5. Estimation of the parameters of the equation. Choose 𝛼1 and 𝛼0 that should
make 𝜇𝑡 as small as possible.
6. Hypothesis Testing – evaluating whether the economic theory is content with
given data, validating or refuting economic theory.
7. Forecasting and prediction
8. Using the model for control and policy analysis.

Concept of Linearity

Assumption of linearity enables the use of the ordinary least squares method. A
mathematical model of the form 𝑌 = 𝑓 (𝑋); 𝑌 = 𝛼 + 𝛽𝑋 − connotes linearity in the
coefficient/parameters and variables.

i. 𝑌 = 𝛼 + 𝛽𝑋 2 + 𝜇
This equation can also be solved using the ordinary least squares. As long as
the coefficients remain linear the assumption of linearity still holds.

ii. 𝑙𝑜𝑔𝑌 = 𝜗 + 𝛿𝑙𝑜𝑔𝑋 + 𝜇

Assumption of linearity still holds

iii. Intrinsically linear regression Model appears non-linear in parameter but can
be transformed into a model linear in parameters e.g. 𝑌 = 𝛽0 𝐿𝛽1 𝐾𝛽2 𝑒 𝜇 - this
is non linear in parameters but after taking logarithms it becomes
ln 𝑌 = ln 𝛽0 + β1 ln 𝐿 + 𝛽2 ln 𝐾 + 𝜇
This makes it linear in parameters but log linear in variables.

iv. Consider the following model 𝑌 = 𝛽0 𝐿𝛽1 𝐾𝛽2 𝜇

ln 𝑌 = 𝛽0 + 𝛽1 ln 𝐿 + 𝛽2 ln 𝐾 + ln 𝜇
This can also be estimated using ordinary least squares techniques.

Intrinsically Non-Linear Models

𝒀 = 𝜷𝟎 𝑳 𝜷 𝟏 𝑲 𝜷 𝟐 + 𝝁

- Cannot be transformed into a linear model using mathematical ways.

- It is impossible to estimate the model using ordinary least squares method.

Page 3 of 16
Types of Data

Time Series Data – collected at regular time interval A time series is a set of
observations on the values that a variable takes at different times. Such data may be
collected at regular time intervals, such as daily (e.g., stock prices, weather reports),
weekly (e.g., money supply figures), monthly [e.g., the unemployment rate, the
Consumer Price Index (CPI)], quarterly (e.g., GDP), annually (e.g., government
budgets), quinquennially, that is, every 5 years (e.g., the census of manufactures), or
decennially (e.g., the census of population).

Cross-Section Data Cross-section data are data on one or more variables collected at
the same point in time, such as the census of population

Pooled Data In pooled, or combined, data are elements of both time series and cross-
section data.

Panel, Longitudinal, or Micropanel Data This is a special type of pooled data in which
the same cross-sectional unit (say, a family or a firm) is surveyed over time.
A panel data (or longitudinal data) set consists of a time series for each cross-sectional
member in the data set. As an example, suppose we have wage, education, and
employment history for a set of individuals followed over a ten-year period. Or we
might collect information, such as investment and financial data, about the same set of
firms over a five-year time period. Panel data can also be collected on geographical
units. For example, we can collect data for the same set of counties in the United States
on immigration flows, tax rates, wage rates, government expenditures, and so on, for the
years 1980, 1985, and 1990.

Review of Probability

Random Variables and Probability Distributions

Probabilities, the Sample Space and random Variables

Outcomes – are defined as the mutually exclusive results of a random process.

Randomness has to do with chance.

The probability of an outcome is the proportion of the time that the outcome occurs in
the long run. E.g. if the probability of your computer not crushing while you are writing
a dissertation is 80%, then over the course of writing many dissertations you will
complete 80% without a crash.

The Sample Space and Events

Sample space – is a set of all possible outcomes

Page 4 of 16
Event – is a subset of the sample space of the sample space i.e. it is a set of one or more
outcomes. The event that my computer will crash more than once is the set consisting of
two outcomes: “no crashes” and “one crash”.

Random Variables

A random variable is a numerical summary of a random outcome. It is a random

mechanism whose outcomes are real numbers. It is a variable whose values cannot be
predicted exactly. Random variables can be distinguished into two a) discrete random
variable and b) continuous random variable.

Discrete Random Variable

Is characterized by the probabilities of outcomes. It is one for which a number of

countable numbers of possible values can occur. It takes only a discrete set of values like
0,1,2 …………….

Continuous Random Variable

A continuous random variable is one for which the set of possible values is some range
or interval of real numbers. It is a variable that takes on a continuum of possible values.

Probability Distribution of a Discrete Random Variable

Simple Linear Regression Analysis

Regression analysis
Regression analysis is concerned with the study of dependency or relationship between
one variable called dependent (regressand/endogenous) variable and another called
independent (explanatory/covariates/exogenous/regressor) variable in order to
estimate and predict the expected value of the dependent variable based on a known
explanatory variable which is fixed in repeated sampling.

A simple linear regression is also known as a bivariate linear regression model. It can be
specified as follows:

y = f ( x)

y = dependent variable and x = independent variable. f ( x ) is a function of x .

The relationship between x and y can be specified in two different types

a) A deterministic or mathematical specification

Page 5 of 16
 b) A stochastic or statistical specification

Deterministic or Mathematical Specification

This is given as follows;

f ( x) =  +  x

y = + x

The values of y can be determined exactly for any given value of x .

For example, given a mathematical specification of the following form:

y = 10 + 0.75 x

This gives the following output

x 0 10 20 30 40
y 10 17.5 25 32.5 40

Stochastic Specification

A stochastic specification also known as an econometric model is the sum of the

deterministic model and the error or disturbance term. This is given as follows:

y = + x+ 

With this specification the values of y for any given values of x cannot be determined
exactly. The error term has a known probability distribution.

Given the equation;

y = 10 + 0.75 x + 
where
 = +50
 = −50

The error has a probability of 0.5. The values of y given x will be as follows:

x 0 10 20 30 40
y (  = −50) -40 -32.5 -25 -15.5 -10
y (  = 50) 60 67.5 75 82.5 90

The simple linear regression model is assumed to be linear and  and  are unknown
regression coefficients or regression parameters.

Page 6 of 16
Relevance of Error Term

Theoretical inadequacy
o Theory itself is inconclusive
o Theory does not specify the number of variables to impact the dependent
variable
Data limitation or inadequacy
Distinction between core and peripheral explanatory variables – attaching
significance to variables.
Inherently random human behaviour.
Poor proxies/measurement errors
Abstraction and parsimony – Occam’s razor i.e. the desire to keep the model as
simple as possible (too much generalization of the model)
Functional misspecification – the effects of using a wrong functional form. This is
captured by the random error.

Criteria for choosing a Good Estimator

1. Unbiasedness – assuming we have an estimator 𝜃 and choose a large sample 𝑛 >

30 and hold 𝑋 fixed. Running it in different samples. 𝐸(𝜃̂) = 𝜃, only valid in
repeated sampling. Suppose we perform repeated sampling whereby we keep
the values of the independent variables constant but change samples of the
independent variable. If we repeat the experiment 3000 times for example, we
then get 3000 estimates of the parameter 𝜃. The value of the estimate of 𝜃 will
change from sample to sample and we thus get a sampling distribution of 𝜃̂ . The
estimator 𝜃̂ is said to be an unbiased estimator of 𝜃 if the mean of the sampling
distribution is equal to 𝜃. i.e. 𝐸(𝜃̂) = 𝜃 . Unbiasedness is only valid in repeated
sampling

Page 7 of 16
 𝑓 (𝜃)

𝑓(𝜃̌)

𝐸(𝜃̌) ≠ 𝜃

2. Efficiency
In some econometric problems we can have a large number of estimators to
choose from. The unbiased estimator whose sampling distribution has the
smallest variance is considered the most desirable of these unbiased estimators.
It is then called the best unbiased estimator or the efficient estimator. NB. It is
mathematically difficult to determine which estimator has the smallest variance.
Econometrics further add the restriction that the estimator had to be a linear
function of the observations on the dependent variable. The best estimator
becomes the best linear unbiased estimator. (BLUE)
3. Minimum Mean Square Criterion
The BLUE criterion allows unbiased to play an extremely strong role in
determining the choice of a good estimator since only unbiased estimators are
considered. An estimator is a minimum MSE estimator if it has the smallest mean
square error defined as the expected value of the squared differences of the
estimator around the true population parameter  .

( ) ( )
2
MSE ˆ = E ˆ − 

The MSE is equal to the variance of the estimator plus the square of its bias.

( ) ( )
MSE ˆ = var ˆ + Bias 2 ˆ ( )
Proof:

( ) ( )
2
MSE ˆ = E ˆ − 

Page 8 of 16
  ( ) ( )  = E ˆ − E ( ˆ ) + E  E ( ˆ ) −   + 2E ˆ − E ( ˆ )  E ( ˆ ) −  
2 2 2
E  ˆ − E ˆ + E ˆ −  
 
But;

( ) ( ) ( ) ( )
2 2
E  ˆ − E ˆ  = var ˆ ; E  E ˆ −   = Bias 2 ˆ
   
And

 ( ) ( ) 
2 E  ˆ − E ˆ   E ˆ −   = 0 because,
  

ˆ − E ( ˆ )  E ( ˆ ) −   = E ˆ E ( ˆ ) −  E ( ˆ ) − ˆ +  E ( ˆ )

2

( ) ( ) ( ) ( )
2 2
  E ˆ  −  E ˆ  −  E ˆ +  E ˆ = 0
   
Therefore;

( ) ( )
MSE ˆ = var ˆ + Bias 2 ˆ ( )

Large Sample Properties of Estimators

Estimators may not possess the good qualities discussed previously. However,
estimators may become good only when the sample size becomes large or when
the sample size tends to infinite thus, we look at the asymptotic or large sample
properties of estimators.

Asymptotic Distribution and Probability Limit

An asymptotic distribution for any estimator is that distribution to which the
sampling tends as the sample size increases. Note that we are not talking of a
single sample that is gradually increasing but we are talking about repeated
sampling when we are taking many samples of fixed sizes.
Assuming we have a random variable 𝑋 and the mean of 𝑋 is 𝑋̅ and the variance
2
of 𝑋 is 𝛿𝑋2̅ = 𝛿 ⁄𝑛
As the sample size increases the variance of 𝑋 → 0.
A zero variance implies no dispersion in the distribution of the mean and the
sampling distribution of 𝑋̅ collapses into a sample value which is equal to the
population mean 𝜇.
When the sampling distribution of an estimator collapses into a single value in
this way the estimator is said to converge in probability to the value. 𝑋̅ is said to
converge in probability to 𝜇 . the value to which it converges to is said to be the
probability limit.
𝑝𝑙𝑖𝑚(𝑋̅) = 𝜇

Asymptotic Unbiasedness
An estimator is asymptotically unbiased if
𝐸(𝜃 ∗ ) → 0 as 𝑛 → ∞

Page 9 of 16
An estimator is said to be asymptotically unbiased if as the sample size increases
the expectation of the mean tends to zero.
Consistency
An estimator is said to be consistent if
𝑝𝑙𝑖𝑚(𝜃 ∗ ) = 𝜃
An estimator is consistent if the sampling distribution collapses on the value of
the parameter being estimated. A consistent estimator has the characteristic that
both the variance and bias of the estimator should tend to zero as the sample size
increases.

𝑀𝑆𝐸 (𝜃 ∗ ) = 𝑉𝑎𝑟(𝜃 ∗ ) + 𝐵𝑖𝑎𝑠 2

𝑀𝑆𝐸 (𝜃 ∗ ) → 0 𝑎𝑠 𝑛 → ∞
Asymptotic Efficiency
An estimator is said to be asymptotically efficient if it has consistency and no
other estimator has a smaller asymptotic variance. An asymptotic efficient
estimator is one whose sampling distribution collapses more quickly to the
parameter being estimated as the size of the sample that are taken increases.

CLASSICAL LINEAR REGRESSION MODEL

Assumptions `

Assumptions of the Explanatory Variable

o Assume that the explanatory variable 𝑋 is non stochastic i.e. its values are
not determined by chance. They are determined by investigation.
o The explanatory variables are fixed in repeated sampling. This implies
that when many samples are taken and we are repeating the experiment
we would choose exactly the same set of 𝑋 vales on each occasion.
∑(𝑋 − 𝑋̅)2⁄ ∑ 𝑥 2⁄
o Variance is 𝑛= 𝑛 → 𝑄 𝑎𝑠 𝑛 → ∞ . where 𝑥 = 𝑋 − 𝑋̅ . 𝑄 is
assumed to b e a fixed finite constant. This implies that as the sample size
increases the variance of 𝑋 does not increase without a limit so we are
ruling out a strong trend in the variable 𝑋.

Assumptions of the Disturbance Term

o We assume that the mean of the error term is zero i.e. 𝐸 (𝜀𝑖 ) = 0
2
o Variance (𝜀𝑖 ) = 𝐸(𝜀𝑖 − 𝐸 (𝜀𝑖 )) = 𝐸 (𝜀𝑖2 ) = 𝛿 2 for all i. we assume that
error terms have the same variance which is 𝛿 2 . If the variances of the
error term are the same, we say we have homoskedasticity and
heteroskedasticity if the variances are different.
o The covariance between any two error terms is zero. 𝐶𝑜𝑣(𝜀𝑖 , 𝜀𝑗 ) =
𝐸(𝜀𝑖 − 𝐸 (𝜀𝑖 )) (𝜀𝑗 − 𝐸(𝜀𝑗 )) = 𝐸(𝜀𝑖 𝜀𝑗 ) = 0 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑖 ≠ 𝑗 (assumption of no
autocorrelation).

Page 10 of 16
 o Each error term is normally distributed with 𝐸 (𝜀𝑖 )=0 and 𝑉𝑎𝑟(𝜀𝑖 ) = 𝛿 2 .
The implication is that there is a large probability of obtaining small
disturbances than obtaining large disturbances. We are more likely to
obtain points close to the population regression function than points far
away from it.

Assumptions concerning the error term and explanatory variable

o Assume that (𝑋𝑖 , 𝜀𝑖 ) = 𝐶𝑜𝑣(𝑋𝑖 𝜀𝑖 ) = 0 . No correlation between the error

term which is stochastic and the independent variables which are fixed
and non stochastic

NB: These assumptions are usually violated when dealing with econometric data.
Regression analysis was developed in a physical laboratory where they used laboratory
experiment. However, when we use real economic data it does not follow the data of
physical sciences and so violations occur. However, the classical linear regression model
provides a good starting point for dealing with differential aspects of regression
analysis.

Derivation of Ordinary Least Squares Estimators

The OLS is the best optimal estimator that minimizes the sum of the squared errors.
Errors are squared to avoid cancellation of positive and negative errors.

Consider the following models

𝑦 = 𝛼 + 𝛽𝑥 + 𝜖 … … … … … … . (1.0)𝑃𝑅𝐹

𝑦̂ = 𝛼̂ + 𝛽̂ 𝑥 … … … … … … … . . (1.1) 𝑆𝑅𝐹

𝜖 = 𝑦 − 𝑦̂ … … … … … … (1.3)

𝑦 = 𝑦̂ + 𝜖 … … … … … … … … … . . (1.4)

𝜖 = 𝑦 − (𝛼̂ + 𝛽̂ 𝑥) ⇒ 𝜖 = 𝑦 − 𝛼̂ − 𝛽̂ 𝑥 … … … … … (1.5)

Square the errors and sum them to get:

2
∑ 𝜖 2 = ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) … … … … … … … (1.6)

In order for us to derive OLS estimators we need to partially differentiate equation 1.6
with respect to the coefficients.

𝜕 ∑ 𝜖2
= −2 ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) = 0
𝜕𝛼̂

−2 ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) 0
= =
−2 −2

Page 11 of 16
 = ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) = 0

= ∑ 𝑦 − ∑ 𝛼̂ − 𝛽̂ ∑ 𝑥 = 0

⟹ ∑ 𝑦 = ∑ 𝛼̂ + 𝛽̂ ∑ 𝑥

⟹ ∑ 𝑦 = 𝑛𝛼̂ + 𝛽̂ ∑ 𝑥 … … … … … … … … … (1.7)

Equation 1.7 is the first normal equation.

̂
Differentiate with respect to 𝜷

𝜕 ∑ 𝜖2
= −2 ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) 𝑥 = 0
𝜕𝛽̂

−2 ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) 𝑥 0
= =
−2 −2

= ∑(𝑦 − 𝛼̂ − 𝛽̂ 𝑥) 𝑥 = 0

⇒ ∑ 𝑋𝑌 − 𝛼̂ ∑ 𝑋 − 𝛽̂ ∑ 𝑋 2 = 0

∑ 𝑋𝑌 = 𝛼̂ ∑ 𝑋 + 𝛽̂ ∑ 𝑋 2 … … … … … … … … … … … (1.8)

Equation 1.8 is the second normal equation.

Solving for 𝛼̂

∑ 𝑦 = 𝑛𝛼̂ + 𝛽̂ ∑ 𝑥

𝑛𝛼̂ = ∑ 𝑦 − 𝛽̂ ∑ 𝑥

Divide throughout by 𝑛 to get;

∑𝑌 ∑𝑋
𝛼̂ = − 𝛽̂
𝑛 𝑛

⇒ 𝛼̂ = 𝑌̅ − 𝛽̂ 𝑋̅ … … … … … … … … … … … … … . . (1.9)

Solving for 𝛽̂

∑ 𝑋𝑌 = 𝛼̂ ∑ 𝑋 + 𝛽̂ ∑ 𝑋 2 ⇒ ∑ 𝑋𝑌 − 𝛼̂ ∑ 𝑋 = 𝛽̂ ∑ 𝑋 2

Page 12 of 16
 ∑ 𝑋𝑌 − (𝑌̅ − 𝛽̂ 𝑋̅) ∑ 𝑋 = 𝛽̂ ∑ 𝑋 2

∑ 𝑋𝑌 − (𝑌̅ − 𝛽̂ 𝑋̅) ∑ 𝑋 = 𝛽̂ ∑ 𝑋 2

∑𝑌 ∑𝑋
∑ 𝑋𝑌 − ( − 𝛽̂ ) ∑ 𝑋 = 𝛽̂ ∑ 𝑋 2
𝑛 𝑛
2
𝑛 ∑ 𝑋𝑌 − (∑ 𝑋 ∑ 𝑌 − 𝛽̂ (∑ 𝑋) ) = 𝑛𝛽̂ ∑ 𝑋 2

2
𝑛 ∑ 𝑋𝑌 − (∑ 𝑋 ∑ 𝑌) = 𝑛𝛽̂ ∑ 𝑋 2 − 𝛽̂ (∑ 𝑋)

2
𝑛 ∑ 𝑋𝑌 − (∑ 𝑋 ∑ 𝑌) = 𝛽̂ (𝑛 ∑ 𝑋 2 − (∑ 𝑋) )

𝑛 ∑ 𝑋𝑌 − ∑ 𝑋 ∑ 𝑌
… … … … … … … … … … (1.10)
𝑛 ∑ 𝑋 2 − (∑ 𝑋 )2

Formular for 𝛽̂ in deviation form

𝑥 = 𝑋 − 𝑋̅ ; 𝑦 = 𝑌 − 𝑌̅

∑ 𝑋𝑌 = 𝛼̂ ∑ 𝑋 + 𝛽̂ ∑ 𝑋 2

∑(𝑋 − 𝑋̅ )(𝑌 − 𝑌̅ ) = 𝛼̂ ∑(𝑋 − 𝑋̅ ) + 𝛽̂ ∑(𝑋 − 𝑋̅ )2

∑ 𝑥𝑦 = 𝛼̂ ∑ 𝑥 + 𝛽̂ ∑ 𝑥 2

∑ 𝑥𝑦 = 𝛼̂ ∗ 0 + 𝛽̂ ∑ 𝑥 2 ⇒ ∑ 𝑥𝑦 = 𝛽̂ ∑ 𝑥 2

∑ 𝒙𝒚
̂=
𝜷 … … … … … … … … … … … … … … (𝟏. 𝟏𝟏)
∑ 𝒙𝟐

Example:

Page 13 of 16
Coefficient of Determination
This measures the proportion of the variation in the dependent variable that can be
attributed to the variation in the independent variable. The coefficient of determination
is given by R2 .

Y
(
Unexplained / residual Y − Yˆ = e )
Yˆ = ˆ + ˆ X

Total (TSS ) )

Explained
Y

TSS = ESS + RSS

Where TSS = Total sum of squares, ESS = explained sum of squares and RSS = residual
sum of squares.

( )
TSS = (Y − Y ) ; ESS = Yˆ − Y ; RSS = Y − Yˆ ( )
Thus;

(Y − Y ) = (Yˆ − Y ) + (Y − Yˆ )
Summing and squaring everything gives;

 (Y − Y ) =  (Yˆ − Y ) +  (Y − Yˆ )
2 2 2

Recall that (Y − Y ) = y , so  (Y − Y ) =  y
2 2
. Therefore’

 y =  (Yˆ − Y ) +  (Y − Yˆ )
2 2
2

Dividing throughout by y 2
gives

Page 14 of 16
 y  (Y − Y ) +  e
2
2 2

=
y
2
y y 2 2

R2

 1 = R2 +
e 2

y 2

The formulae for the coefficient of determination is therefore given as follows

R 2
= 1−
e 2

y 2

Variant formulas for R2

ˆ 2  x 2 ˆ 2 S x2 ˆ  xy
R2 = = =
y 2
S y2 y 2

Given the following data

Price (X) 2 7 5 1 4 8 2 8
Output(Y) 15 41 32 9 28 43 17 40

a) Estimate the bivariate linear regression model with price as the independent
variable.

b) Compute the residuals for the data and hence calculate e 2

c) Calculate the coefficient of determination and comment.

Correlation Coefficient

Correlation coefficient measures the strength of relationship between variables and is

calculated as:

n XY −  X  Y
r=
( n X 2
)(
− (  X ) • n Y 2 − (  Y )
2 2
)
In deviation form it is calculated as follows:

r=
 xy
( x) • ( y)
2 2

The correlation coefficient lies between -1 and 1. A positive correlation coefficient

indicates that variables moves together.
Page 15 of 16
Covariance between ̂ and ˆ

( ) (
Cov ˆ , ˆ = E ˆ − E ˆ ( )) ( ˆ − E ( ˆ ))
Recall, ˆ = Y − ˆ X , so the covariance is given as follows,

( ) (
Cov ˆ , ˆ = E Y − ˆ X − E ˆ( )) ( ˆ − E ( ˆ )) ;
E (ˆ ) = E (Y − ˆ X ) = Y − XE ( ˆ )
E (ˆ ) = Y − X 
  2 
( ) (  ) ( ) ( ) ( )
2
cov ˆ , ˆ = E − X ˆ −  ˆ −  = − XE ˆ −  ˆ
= − X var  = − X 
  x 2 
 

The variance of ˆ is always positive and so the covariance depends on the sign of X . If X is
positive then the covariance will be negative.

Hypothesis Testing
Refer to your Biometry lessons

There are two approaches for performing hypothesis testing

1) Confidence interval approach

2) Test of significance approach

Using the previous data and findings

a) Using the 5% level of significance, test the hypothesis that price influences output.
b) Construct the 95% confidence interval for the two coefficients
c) Calculate the covariance between the two coefficients and comment
d) Calculate the correlation coefficient and comment

Page 16 of 16