代写stata编程、代写stata语言、Empirical Methods in Microeconomics

日期：2018-04-19 10:30

Problem Set #4 for  EC439 – Empirical Methods in Microeconomics

This problem set should be done entirely in Stata.

Please email your .do files to uoft.ec439@gmail.com before April 18th, 2018. Late

papers will be docked 5x % per day where x is the number of days late (since this is

so close to my grades deadline I really need them in by then). Weekend days count

as days, and the time on the email will count as the time submitted. New days start

at 12:00 midnight.

Include the explanation for each answer as a comment below the code using the /*

COMMENT */ tags in Stata.

For this assignment we’re using the data posted on the blackboard site for the

course. The data is from 2 related papers by Fabian Waldinger and are testing

material from lectures on panel data and fixed effects models.

Both papers look at the expulsion of high profile scientists from Germany by the

Nazis in the early 1930s. Both estimate the effects that this had on the people

remaining in Germany, so both are papers about “peer effects” in productivity.

Dismissals were based entirely on family history, religion and political views and

can be thought of as strictly exogenous.

Data set #1:

We’ll use the following variables in this data:

phduni: university where student graduates year of graduation

phdyear: year of graduation

female: female indicator

foreign: foreign indicator

Outcomes:

ordprof : indicator whether student becomes full professor

publishedtop : indicator whether student publishes dissertation in top journal

lifetimecit: number of lifetime citations of former student

Faculty measures:

avg_quality: average faculty quality measured by citations in top journals

cohort_profs: students per professor (all mathematics students in department)

change_avg_quality  measures how much faculty quality changes due to dismissal.

0 for all pre dismissal years and equal to quality change thereafter (remains 0 for

departments without dismissals).

change_cohort_profs : measures how much students per professors change due to

dismissal

phduni331 – phduni3333 a dummy which is equal to 1 for the university that the

PhD student graduated from (if she graduated before 1933) or for the university

that she attended at the beginning of 1933 (if the student graduated after 1933).

Before starting a project it’s good practice to save your control set as a macro so that

your specifications are consistent throughout and your code is transparent. Using

the global command set a control set to include phd year fixed effects, father

occupation fixed effects, the female and foreign variables and the phduni33

variables.

Question 1 (out of 15): For this question use dataset #1, described above.

a) We want to test whether the change in faculty quality induced by the Nazi

dismissals reduced the productivity of those that were not dismissed. First, regress  

each of the three outcomes on the change in department quality and our controls.

What can you conclude about productivity spillovers?

b) Explain how this model can be thought of as a difference-in-differences model.

c) What do you think of the standard errors in part (a)? Do you think we should

cluster them and why? If we did cluster standard errors, at what level should we

cluster? What happens if you do this? What does this say about the observations?

Question 2 (out of 15): For this question use dataset #2. We only need 3 variables

for this question:

Lcit50_peersub_avg_ is the peer quality in a persons department

pub_ is the persons productivity as measured by publications

uni_ is the university that the person belongs to

a) Regress the number of publications (pub_) on a measure of peer quality at the

department level (Lcit50_peersub_avg_). Is there a positive effect of peer quality on

output? Can we interpret the relationship as causal?

b) Somebody suggests we manually compute the within estimator (based on uni_).

Manually program the within-effects estimator for them that controls for university

fixed effects. How do the estimates change and why do you think the estimates

change? What words of warning would you give them about either the estimates or

the standard errors that they’ll get using your program?

c) We know there’s a better way. Estimate the same model using the dummy

variable method. Are the estimates different? How about the standard errors? Why

do we see any differences?

2 Bonus points: Fewer than 20 lines of code overall.