Name:…………………………………………. ID No:…………………………..

BITS-Pilani Dubai Campus

ECON F241 ECONOMETRIC METHODS
Semester I, 2018Test-1 (Closed Book)
Section 1: Multiple Choice Questions (1X11)
Note: Encircle the most appropriate option and return this paper with answer book!
All the notations have usual meanings unless otherwise defined!
1 Econometrics is different from any field of pure science because it relies on
a) Experimental data b) Non-experimental data c) Controlled data
2 Suppose you are estimating the relationship between price P and the quantity demanded
for a particular commodity X. Which of the following is the most appropriate
econometric specification
P 100 90 80 70 60 50
X 0 5 10 15 20 25
a) P = β1- β2Xwhere,-3< β2<-1b) P =-β2X where,-3< β2<-1
c) P = β2X where, 1< β2< 3d)P = β1 + β2X where, 1< β2<3
3 Assume that you want to examine the relationship between income (Y) and
consumption spending (C) of the consumers living in UAE at a given point of time.
Which of the following regression model will you use
a) Yi= β1 + β2Ci + ui b) Yi= β2Ci + uic) Yi= β1 - β2Xi + ui
4 The property of sample regression function that the E(ui)=0, implies that the regression
model Yi= β1 + β2Xi + ui can be alternatively written as
a)
b)
c)
5 Assume a regression model Wi = βˆ1- βˆ2Ci + βˆ3 Ii + Ui where covariance of Ui and Ci
is Not equal to zero. Then β2measures the impact of
a) C on W b) C and U on W c) C and I on W d) Can not say
6 The Classical Regression model involves estimation of relationship between
a) Dependent non-random variable and independent random variable(s)
b) Dependent random variable and an independent non-random variable(s)
c) Dependent random variable and an independent random variable(s)
d) Dependent non-random variable and an independent non-random variable(s)
7 The property that there is zero correlation between dependent variable Y and residual
series (u) obtained from using OLS, implies that the
a) u generated is purely random b) u does not reflect the systematic effect of any
omitted variable c) u denotes random changes in Y d) All, a, b and C
8 The regression model: Yi= β1 + β2Xi + ui implies that the actual Yi consists of
a) deterministic component b) stochastic component c) random component
d) both a and b e) both b and c
9 Suppose in a sample regression line Wi = βˆ1- βˆ2Ci + Ui , the variance of Ui increases
 as C increases and the heteroscedasticity in errors. Then it this implies the reliability of
Y observations will
a) increase as C increases b) decreases as C increases c) remain same regardless of C
10 Suppose, you estimated a hypothetical sample regression function using OLS method
and you observed that  uˆi2 = (Yi – Y^i)2 = 0. What does it imply
a) βˆ1 , βˆ2 are wrong estimates for β1 , β2
b) βˆ1 , βˆ2 are not close approximation for β1 , β2
c) βˆ1 , βˆ2 are perfect/exact approximation for β1 , β2
d) βˆ1 , βˆ2 are close approximation for β1 , β2
11 Suppose, you estimated a regression model using OLS technique: Q= 2.25-0.7P +
0.2Y; where P is the price and Q is the spending ona certain Commodity and Y is
income. What does the coefficient w.r.t. Y indicate
a) If income increases by 1 dollar, the spending on the commodity will increase by 0.2
dollars, on average
b) If income increases by 1 dollar, the spending on the commodity will decrease by 0.2
dollars, on average
c) If income increases by 1 dollar, the spending on the commodity will increase by 0.7
dollars, on average

Section 2: Descriptive Question (3X3)

15.State and explain the various stages of estimating any econometric relationship.

16. Explain the various properties of Sample regression line obtained by using Ordinary Least
Square technique.

17. Explain the significance of disturbance term in stochastic specification of population

regression function?

***********************
 Name:…………………………………………. ID No:…………………………..
BITS-Pilani Dubai Campus
ECON F241 ECONOMETRIC METHODS
Semester I, 2018 Test-2 (OPEN Book)
Section 1: Multiple Choice Questions (1X10)
Note: Encircle the most appropriate option and return this paper with answer book!
All the notations have usual meanings unless otherwise defined!
1 The probability of commenting an error that is ‘rejecting the null hypothesis which is
actually true’ is given by
a) p-value b) 1-(p-value) c) both a and b d) t-value e) s.e.
2 A higher variation in independent variable will result in
a) Lower s.e b) higher s.e c) will have no impact on s.e
3 Assume a multivariate regression model given by Yi = βˆ1- βˆ2Xi + βˆ3 Fi + βˆ4 Zi + Ui
.The degrees of freedom of this model is measure by
a) n - 1 b) n - 2
c) n- 3 d) n-4
4 Suppose you estimated a regression model Wˆi = βˆ1- βˆ2Ci + βˆ3 Ii . Assume the exact
level of significance that is p-value associated with coefficient βˆ2 is 0.049. Setting
conventional level of significance, which of the following inference will be more
appropriate
a) Reject the Null hypothesis b) Accept the Null hypothesis c) Can’t say anything
5 One implications of Gauss-Markov theorem is that an OLS estimator is Efficient. This
efficiency of an estimator implies that
a) the likelihood of an OLS estimator being equal to true population is high
b) the likelihood of an some other estimator being equal to true population is high
c) the likelihood of the mean of OLS estimator being equal to true population is low
d) the likelihood of mean of some other estimator being equal to true population is high
6 Suppose you want to estimate the relationship between number of years of experience
(E, measured in terms of number of years employed) and salaries of employees (S,
measured in terms of AEDs) of a particular industry, and you set a regression model:
Si = βˆ1+βˆ2Ei + Ui. Which of the following, you think, will be most appropriate null
Ho and alternative H1 hypothesis
a) Ho : βˆ2=0; and H1: βˆ2≠0 b) Ho : βˆ2=0; and H1: βˆ2>0
b) Ho : βˆ2=0; and H1: βˆ2<0 d) Ho : βˆ2=0; and H1: βˆ2>-0.45
7 A measure of coefficient of determination is different from that of correlation
coefficient, because coefficient of determination is
a) a linear measure b) a symmetric measure c) an asymmetric measure
d) both a and b e) all a, b and c
8 The Normality assumption of CLRM implies that the
a) βˆ2 is normally distributed b) Yiˆ is normally distributed
c) Uiˆ is normally distributed d) All
9 We know that a country with high GDP today are going to have higher GDP tomorrow
as well. Suppose you are interested in examining the factors that determine the GDP
level (Y) of a country. Theoretically, the main factors which may effect GDP include
investments on Land (L) Labor (N), Capital(C) and Entrepreneurship (E). And now
 suppose that you used a regression model: Yt = βˆ1 + βˆ2 Lt + βˆ3 Nt + βˆ4 Ct + βˆ5 Et +
Ut instead of: Yt = βˆ1 + βˆ2 Lt + βˆ3 Nt + βˆ4 Ct + βˆ5 Et + βˆ6 Yt-1 + Ut. Which
assumption(s) of CLRM you suspect might be violated
a) Linearity b) No-Autocorrelation c) Homoscedasticity
d) Multicollinearity e) All
10 Assume that you want to estimate the relationship between time spend in gym (T) by
students and their Fitness (F). Also assume that there is maximum limit of time one can
spent in gym and anymore time spend beyond that gives more muscle stress and injuries
and reduces fitness levels. As this relationship is inverted U-shaped, you specify it as
F = β1 +β2T - β3 T 2+ u. Now, what would you expect r and r-square to be
a) r-square > 0 and r = 0 b) r-square = 1 and r = 0
c) r-square = 0 and r > 0 d) r-square < 1 and r < 0

Section 2: Descriptive Question

11. Explain the concept ofAutocorrelation. What problem does the presence of it cause for the
OLS estimator? (3)

12. Define the Type I and Type II error. (2)

13. State and explain the various statistical properties of an OLS estimator? (2)

14. State and explain the four assumptions of the classical linear regression model? (3)

***********************
 BITS-Pilani Dubai Campus
F241 ECONOMETRIC METHODS
Semester I, 2018 Comprehensive Examination (Closed Book)
Time Allowed: 03:00 hrs Marks: 60 Date: 19/12/2018
Very Short Answer type questions: Answer all the questions (5X2)

1. Define a pure white noise time series process.

2. State one main difference between Statistical relationship and Deterministic relationship.
3. State two features of pure random walk process.
4. Define the term Homoscedasticity.
5. Distinguish between Controlled and Uncontrolled data. State and explain the various types of
uncontrolled data?
Short Answer type questions: Answer all the questions (6X5)

6. Define a non-stationary process. State and explain various types of non-stationary time series
processes.
7. State and explain the two approaches for testing the hypothesis.

8. Suppose the DGP of time series variable (say output) is given by yt  0  1t  2 yt 1  ut ,
where 0  0; 1  0;  2  1 . Show that its mean and variance are time variant.

9. Explain the Augmented Dicky-Fuller (DF) test for testing the Unit root of a time series
process.

10. Assume a linear regression model Yi  0  1 yi  ui which is estimated using OLS method.
State and explain the properties of  0 and 1 under the assumption that ui follows a normal
distribution ui : N (0,  2 ) .

11. Define the term Autocorrelation. Explain the consequences of autocorrelation on OLS
estimates.
Essay Type Questions: Answer ALL questions (2 X10)

12. Suppose that you are interested in examining the factors that determine Agricultural output
overtime in an economy. Suppose you estimated a regression model:
Yt  0  1 X t  2 Lt  3 M t   4Wt  ut using OLS method, where, X is Arable Land used, L is

labor used, M is machinery/technological factors and W is average temperature (weather)

recorded. The results are summarized in the table given below.

Coefficient s.e. t-value p-value r-square

 0 0.01 0.11 0.09 0.80 0.85
1 0.31 0.04 7.75 0.01
2 0.52 0.096 6.41 0.00
3 0.13 0.21 0.61 0.70
4 0.10 0.04 2.50 0.04

Interpret these results: (i) What does these results indicate about the relationship between Y
and X, W, M, L.

(ii) Evaluate the null-hypothesis that i  0 based on p-values and/or t-test statistic.

(iii) Which factor (Labor, Machinery, Land, and Weather in terms of productivity) is more
important in this production process?

(iv) What does r-square indicate?

13. (i) Define a hypothesis. Using an appropriate example, construct a Null and an Alternative
hypothesis for examining a relationship. (ii) State and discuss the various
conditions/assumptions which are necessary to suffice the BLUE property of OLS estimators.