Scribd
Upload
16 views2 pages

PS09 2023

This problem set focuses on functional form misspecification, measurement errors, and regression analysis using data from the NLS young men cohort. It includes questions on the RESET test, consequences of measurement errors, and various regression analyses related to education and wages. Students are required to replicate findings from David Card's influential paper and analyze how different variables affect the coefficients in their regressions.

Uploaded by

Aryan Verma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views2 pages

PS09 2023

This problem set focuses on functional form misspecification, measurement errors, and regression analysis using data from the NLS young men cohort. It includes questions on the RESET test, consequences of measurement errors, and various regression analyses related to education and wages. Students are required to replicate findings from David Card's influential paper and analyze how different variables affect the coefficients in their regressions.

Uploaded by

Aryan Verma
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Econometrics: Problem Set #9

Abdul G Noury
7th December 2023, 11:59 PM

Problem 1
ˆ What is functional form misspecification? Give some examples to answer this
question.

ˆ Explain how the RESET test can detect a functional form misspecification.

ˆ In a regression equation log(y) = β0 + β1 x1 + β2 x2 + β3 x3 + β4 x4 + β5 x5 + u


we had R2 = .353 (n=177). When we also included x24 and x25 we obtained
R2 = .375. Is there evidence of functional form misspecification in this model?

ˆ How can we test against non-nested alternatives (using Davidson-MacKinnon


test)? What are some limits of this approach?

.........

Problem 2
ˆ Explain what we mean by measurement errors, and which G-M assumption is
likely to be violated in presence of measurement errors?

ˆ What are the consequences of measurement error in independent variable (x1 =


x∗1 + e1 )? Consider two cases: (i) cov(x1 , e1 ) = 0 and (ii) cov(x∗1 , e1 ) = 0.

ˆ What would be some solutions to measurement error problem?

.........

1
Problem 3 2

Problem 3
In this problem set we use data from the NLS young men cohort, collected by David
Card and used in his influential paper“Using Geographic Variation in College Prox-
imity to Estimate the Return to Schooling”. The dataset (Card95) is posted on
Brightspace. Carefully read the paper, Card (1995), and try to replicate the main
tales: Table 1, 2 and 3 (no points for this replication, only the following sub questions
carry points.).

ˆ Describe and summarize the data. Some important things to check out are the
age distribution and the distribution of key variables.

ˆ Run a regression of log wage on education.

ˆ Add potential experience and it’s square to the regression. How does the coef-
ficient on education change? Why does the coefficient change in that direction?

ˆ Add the “iq” variable to the regression. Notice how the “ed76” coefficient
changes (why?).

ˆ Add the kww variable. This is a test score from a non-cognitive survey exam-
ining how much young men know about the work force. Notice how the “ed76”
coefficient changes and how the iq coefficient changes. Briefly explain.

ˆ Use nearc4 to instrument for education in predicting log wages (exclude kww
and iq from the regression). What are the income returns to education according
to this regression (they will be large).

ˆ Add controls for locational dummies to the above regression. The locational
dummies are 9 region dummies and an indicator for smsa: reg661-reg668 smsa66r.
Note how the coefficient on education changes as a result of controlling for lo-
cation. How does adding locational dummies change the assumption necessary
for the nearc4 instrument to be valid? Is nearc4 a plausible instrument in this
context? Test for relevance.

ˆ Industry is a major determinant of wages. Should we control for industry in


trying to determine the return to education? Why or why not?

.........

Econometrics PS #9

You might also like