Scribd
Upload
156 views

Econometrics II

This document contains instructions and questions for a mid-term examination in Econometrics II. It is divided into three sections. Section I covers limited dependent variable models, including differences between logit and probit models. Section II addresses simultaneous equation models, identifying endogenous and exogenous variables. Section III discusses panel data models, comparing pooled, fixed effect and random effect methods. Students are instructed to show all work and attempt all questions over four pages within a two hour time limit.

Uploaded by

Nia Hania Solihat
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
156 views

Econometrics II

This document contains instructions and questions for a mid-term examination in Econometrics II. It is divided into three sections. Section I covers limited dependent variable models, including differences between logit and probit models. Section II addresses simultaneous equation models, identifying endogenous and exogenous variables. Section III discusses panel data models, comparing pooled, fixed effect and random effect methods. Students are instructed to show all work and attempt all questions over four pages within a two hour time limit.

Uploaded by

Nia Hania Solihat
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

TAZ.04-SOP.01/F.

02

Institut Agama Islam Tazkia


Mid-Term Examination
First Semester 2021/2022
SUBJECT / CODE : Econometrics II
LECTURER : Dr. Indra, S.Si, M.Si
DAY / DATE : Senin/8 November 2021
TIME DURATION : 2 HOURS (13.00-15.00 WIB)
ROOM : Online

INSTRUCTIONS TO CANDIDATES

1. CANDIDATES ARE REMINDED THAT THEY MUST NOT HAVE BOOKS, NOTES,
PAPER OR OTHER MATERIAL IN THEIR POSSESSION UNLESS THEIR USE IS
SPECIFICALLY PERMITTED BY "INSTRUCTIONS TO CANDIDATES" SET OUT
BELOW.

2. This paper consists of 3 questions printed on 4 pages.


PLEASE CHECK BEFORE COMMENCING. This is a MID-TERM Examination.
3. Attempt all Questions.
4. This paper is worth a total of 35 marks.
5. Calculators are permitted.
6. Candidate must not remove this paper from the examination room

I. LIMITED DEPENDENT VARIABLE MODEL


a. Describe the differences in characteristics between the limited dependent
variable model and the common regression model!
b. Describe the advantages of the Logit model and the Probit model in estimating
models with qualitative (dichotomy) dependent variables compared to the Linear
Probability Model (LPM)!
c. Describe the difference between the Logit Model dan the Probit Model!
d. A study observes factors affecting MSMEs (UMKM) decisions in Bogor City to
apply for Halal Certification. Suppose this study employs a Logit regression
model which is expressed as follow:
Yi =  o + 1 X 1i +  2 X 2i + 3 X 3i +  4 X 4i +  i
where
• Y= 1 if an MSME carry out Halal Certification, and Y = 0 if otherwise
• X1 is MSME profit (in million Rupiah)
• X2 is the education level of MSME owner (1 if graduated from university,
and 0 if otherwise)
• X3 is the number of employments (people)
• X4 is the number of MSME products (units)

Page 1 of 4
TAZ.04-SOP.01/F.02

The estimation result of logit regression model is provided as follow:

Variable Coefficient P-Value.

C 3.760 0.031
X1 1.203 0.035
X2 1.160 0.072
X3 0.221 0.113
X4 1.305 0.056

Based on the estimation result above:


i. Determine factors that have a significant effect on MSMEs (UMKM)
decisions in Bogor City to apply for Halal Certification (note: use 10%
significant level)? Explain your answer!
ii. Determine the odds ratio of the variables that have a significant effect and
give an interpretation!

II. SIMULTANEOUS EQUATION MODEL


a. Explain the following terms relating to the simultaneous equation model and give
an illustration:
i. Endogenous variable
ii. Exogenous variable
iii. Endogeneity problem
iv. Structural equation
v. Identity equation
vi. Reduced Form equation
b. Consider the following simultaneous equation model:
Y1t = 10 +12Y2t+13Y4t+β11X1t++β12X2t +e1t
Y2t = 20 +23Y3t+24Y1t +β21X1t+β22X3t+e2t
Y3t = 30 + 31Y1t +β31X1t+ β32X2t+ β33X3t +e3t
Y4t = 40+41Y1t +42Y2t +43Y3t+ β41X2t + β42X4t + e4t
Based on the simultaneous equation model above:
i. Identify endogenous and exogenous variables.
ii. Use the order condition and the rank condition to identify whether each
structural equation is under-identified, exactly identified, or over-identified!
iii. What method do you think is suitable for estimating the equation parameters
above? Explain your answer!

III. PANEL DATA MODEL


a. Explain the advantages of using a panel data model compared to pure time
series and pure cross-section model!
b. Explain the attrition of the panel data model!

Page 2 of 4
TAZ.04-SOP.01/F.02

c. Explain the difference between the pooled least square estimation method, the
fixed effect method, and the random effect method in the panel data model!
d. Consider a specification of panel data regression model as follow:
ln Yit =  o + 1tit +  2 ln Lit +  3 ln K it +  it
where
i. lnY denotes the natural logarithm of the output
ii. t deniotes time trend
iii. lnL denotes the natural logarithm of the number of labor
iv. lnK denotes the natural logarithm of the number of capital

Suppose the model has been estimated (with Eviews Program) using three
approaches, namely: pooled least square, fixed effect method, and random effect
method. Then the selection of panel data model was tested, and the results were
as follows:

Chow F-test:

Effects Test Statistic d.f. Prob.

Cross-section F 2.041879 (139,889) 0.0025


Cross-section Chi-square 285.659338 139 0.0021

Hausman test:

Correlated Random Effects - Hausman Test


Equation: Untitled
Test cross-section random effects

Chi-Sq.
Test Summary Statistic Chi-Sq. d.f. Prob.

Cross-section random 83.194546 2 0.015

Cross-section random effects test comparisons:

Variable Fixed Random Var(Diff.) Prob.

CAP -0.000941 -0.001008 0.000007 0.9799


EMP 0.011463 0.000382 0.000002 0.0000

LM Test test:
Lagrange Multiplier Tests for Random Effects
Null hypotheses: No effects
Alternative hypotheses: Two-sided (Breusch-Pagan) and one-sided
(all others) alternatives

Test Hypothesis
Cross-section Time Both

Page 3 of 4
TAZ.04-SOP.01/F.02

Breusch-Pagan 3322.456 40796.38 44118.84


(0.0013) (0.0112) (0.0134)

Based on the test results above, which model should you choose? Explain your answer!

Page 4 of 4

You might also like