SHYAM LAL COLLEGE LIBRARY (M)

[This question paper contains 28 printed pages.]

Your Roll No.....*

Sr. No. of Question Paper : $213 H

Unique Paper Code : 2272102403

Name of the Paper : Introductory Econometrics
Name of the Course :B.A. (Hons.) Eeonomies (NEP-UGCF-2022)
DSCC-12

Semester : IV

Duration: 3 Hour Maximum Marks: 90

Instructions for Candidates

1. Write your Roll No. on the top immediately on receipt of this question paper.
Question 1 is comnpulsory. Answer any four questions from the rest of the six
2.
questions. In total, five questions to be answered.

3. Use of simple caleulator is permitted.

4 Allintermediate calculations should be rounded off to four decimal places. The
values provided in statistical tables should not be rounded off. All final caleulations
should be rounded off to two decimal places.
5 Answers may be written either in English or Hindi; but the same medium should
be used throughout the paper.

SHYAM LAL COLLEGE LIBRARY (M)

PT.O.

SHYAM LAL COLLEGE LIBRARY (M)

5213 2
52
State whether the following statements are true or false. Give reasons for your
answer

(a) For an OLS estimated regression cquation, ,=b, +b,X, the sum of the
product of the residuals and mean deviation of the cxplanatory variable is
zero.

(b) Ceteris paribus, the higher the VIF (Variance Inflation Factor), the smaller
is the variances ofOLS estimators.

(c) For an OLS estimated regression equation, , =b, + b,X, if we multiply

both Yand X by 1000 and re-estimate two variable regression model, the
intercept coefficient will increase by 1000 times.

(d) For a three variable regression model, TSS is always equal to the sum of
ESS and RSS.

(e) It can casily happen that F statistic is significant while t statistics of

explanatory variables are not significant.

() In a regression model with intercept, number of dummies for each

qualitative variable must be one less than the number of categories of that
variable. (3×6)

() syaa (oLS) fr urfR ufavr t =b, + b,X,

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)

$213

() f ra (Ceteris paribus), VIF (w Trifa HE) fr JR A,

() shyeçe fAr fAR sfaTHT TAGM = b, +byX, t, t a Y

2 Using data for 110 schools the following model of monthly expenditure incurred
by a school was estimated :

log(?), = 138.8- 1.571 D, + 0.808 X-0.0944 Dx,

(2.864) (-106) (13.466) (-2.04)

Y annual expenditure incurred by a school, in thousands of rupees X= number

y government school
school DiJ
of students in the ifprivate school
(i) Write the regression equation for government school and private school
separately. Also, interpret the slope coefficient of two categories.
SHYAM LAL COLLEGE LIBRARY (M)
PT.O.

SHYAM LAL COLLEGE LIBRARY (M)

$213
4

(n) Check the statistical sipnifcance of dafferental intercept cocfficicnt and slope
dilier at 10% level of siynificance. Based on the conclusion of thesc lests,
draw the population reyression lines for the governmcnt and private schools.

(m) Test the hypthesis that the population error term is nornally distributed at
1% level of significance when it is given that for the residuals, skewness is
0.2% and the value of kurtosis is 2. Also, in classical linear regression model,
what is the rationale for the assumption that population error term follows
nornal distribution?
(6,6,6)

log(Y), = 138.8- 1.571 D, + 0.808 X, -0.0944 D,X,

(2.864) (-1.06) (13.466) (-2.04)

X = A fuft

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)
$213 5

3 (a) For estimating the Phillips curves for the United States from 1958 to 1969
the following regression was obtained

9. = -02594 20.5880
te (-02s72) (4.3996)
R²=0.6594, d = 0.6394

Y= Rate of change of money wages in percent

X = Unemployment rate in percent

(i) Interpret the intercept estimate. Is there any evidence of first order
autocorrelation in the residuals at 5% level of significance?

(ü) Outline the method of estimation that will produce BLUE estimators in
the presence of first order autocorrelation.

(b) Consider the following regression function on sales revenue for a particular
firm for last 10 months, as estimated by OLs

SHYAM LAL COLLEGE LIBRARY (M)

PT.O.

SHYAMttAt COttEGE ttBRARY(M)

6
$213 5213

P= 51508 + 0.054 X,
se = (4.03) (0.04)

&= 1.02
T2S X=6500

Y is the monthly sales revenue in billions of dollars

X is the monthly expenditure on advertising in millions of dollars.

Find the predicted mean sales revenue if the advertising expenditure for the
99% confidence
firm in the next month is 30 million dollars. Also, find the
interval for the truc predicted mean of sales revenue corresponding to 30

million dollars advertising expenditure. (6+6,6)

() 1958 1969 6 g PR

.=-02594 + 20.s880
t (-02572) (43996)
R0.6594, d =0.6394

4. In

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)
$213 7

9,= 51.508 + 0.054 X,

se =(4.03) (0.04)

=1.02
I-25 SX 6500

Y ffer tR fe R

4. In a regression of housing expenditure in rupees (Y) on annual income of

families in rupees (X) for a sample of 27 familics, the following results were
obtained :

SHYAM LAL COLLEGE LIBRARY (M)

PT.O.

SHYAM LAL COLLEGE LIBRARY (M)

$213

Variable Coefficients Standard Eror

0.121 0.009
Constant 3.803 4.570
n-27
R= 0.776

On plotting the residuals against X, it was found that the variance of the residuals
increased with X,

() What problem does this indicate? Name any one test for its detection and
explain the steps to conduct this test.

(ü) What are the consequences of this problem for OLS estimators? Which type
of dataset is more likely to be characterized by this problem?

(ü) Explain the estimation process of Weighted Least Squares with

unknown error variances in this context where error variance is
proportional to X*. Also, mention which is the intercept and slope
parameter in the transformed model and how do we get back to the original
model. (6,6,6)

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)
$213 9

X 0.121 0 009

B803 4 570

27

R0776

(a) Suppose a third-degree polynomial regression was fitted to a cost-output

model for 18 firms and the following results were obtained :

SHYAM LAL COLLEGE LIBRARY (M)

PT.O.

SHYAM LAL COLLEGE LIBRARY (M)

5213 10

Cost, = 141.7667 +0.208 Q, - 0.086 Q+0.0054 Q

se= (6.37) (4.78) (2.98) (6.98)

Qis the output of the i firm.

(i) If cost curves are to have U-shaped average and marginal cost curves,
then what are youra priori expectations about the intercept and slopc
estimators? Check if these a priori expectations are satisfied for the
given model.

(ii) Interpret the estimated slope coefficient of Q' and test its significance
at 10% level of significance.

(b) Let Y be the Gross National Product, X, be the Exports, X, be the Imports
and X, be the Net Exports in the following relationship:

Y B, + B,X, + B,x, +B,X, + U,

Which assumptions of Classical Linear Regression Model is violated here?
If you estimate this equation by ordinary least squares, then which of the
parameters can be estimated? Explain. (6+6,6)

Cost, = 141.7667 + 0.208 Q, - 0.086 Q + 0.0054 Q

se= (6 37) (4 78) (2 98) (6 98)

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)
5213

() frr Y v g JR t, x, Prta , rTA X, s Arfafea

u Aufa Xx, 8:

Y, - B, + B,X, + B,X, + B,X, + U,

6
(a) For a particular cafeteria, the following equation was estimated using yearly
data on cups of chocolate flavoured cold coffee sold:

log(Q.) = 1.534 - 0.250 log (P) +0.025 log(P:")

se = (0.2001) (0.240) (0.016)
R² =0.804 TSS = 4803

Where

Q,- cups of chocolate flavoured cold coffee sold

P, price (in rupees) per cup of chocolate flavoured cold coffee
P"- price (in rupees) per cup of vanilla lavoured cold coffee
t - 199-2023

SHYAM LAL COLLEGE LIBRARY (M)

PT.0.

SHYAM LAL COLLEGE LIBRARY (M)

5213 12

Note that all cups in the cafeteria are of same size.

(0) What can you say about the own-price and cross-price elasticity of
demand for chocolate Navoured coffee? Test the hypothesis that the
own price elasticity of demand of chocolate flavoured coffee is unitary
elastic at 10% level of significance.

Draw the ANOVA table for the regression cquation. Also calculate R'.

(b) Explain the statement "It is better to include irrelevant variables than to
exclude relevant ones". (6+6,6)

log(@:) = 1.534 0.250 log(P.) +0.025 log(Pe*)

se = (0.2001) (0.240) (0.016)
R² = 0.804 TSS = 4803

P" = fa ()

t = 1991-2023

SHYAM LAL COLLEGE LIBRARY (M)

 SHYAM LAL COLLEGE LIBRARY (M)

$213

(a) Explain the steps involved in performing Ramsey's RESET test for model
specification,

(b) Using data collected from 1990 to 2022, for a particular country, the following
cquation was estimated using OLS :

9. = 4.47 + 0.34 log X -122 log X

( (4.28) (5.31) (-0.98)
R² = 09987

Y -Child Mortality Rate

X,-Female Literacy Rate
X,- Per Capita GNP

() Are the regression results in conformity with your a priori expectations?

If not, what can be the reason behind it? Explain any three remedies
to overcome this problem.

SHYAM LAL COLLEGE LIBRARY (M)

SHYAM LAL COLLEGE LIBRARY (M)

5213 14

(0) Now suppose the researcher reeressed female literacy rate on per
capita GNP and in this second regression, the R²is 0.5284. Do you
think that the reason that you suggested in (i) is significantly present in
the model, at 5% level of significance? If your answer is yes, then do
you think that the presence of this is necessarily bad? (6,6+6)

)f¿t fty fR 1990 2022 as T ta Ja

9=447 + 0.34log X¡i - 132gX

t- (4.28 (5.31) (-0.94)
R² =0.9987

Y = R

X, = sRaT

X, = yf fe GA JeTE (yrt)

SHYAM LAL COLLEGE LIBRARY (M)