ECA5103 Assignment 3
All students have to hand in this assignment by 5pm Monday 22 November 2019. Please
submit both your answers and Stata code).
1. Use the state-level data on murder rates and executions from http://fmwww.bc.edu/ec-
p/data/wooldridge/murder.dta for the following exercise.
(a) Consider the unobserved effects model
mrdrteitit = ηt + β1execit + β2unemit + ai + uit
where ηt simply denotes different year intercepts and ai is the unobserved state effect. If
past executions of convicted murderers have a deterrent effect, what should be the sign
of β1? What sign do you think β2 should have? Explain.
(b) Using just the years 1990 and 1993, estimate the equation from part (i) by pooled OLS.
Do you find any evidence for a deterrent effect? Comparing the results with and without
the cluster(id) option.
(c) Now, using 1990 and 1993, estimate the equation by fixed effects. You may use first
differencing since you are only using two years of data. Is there evidence of a deterrent
effect? How strong?
(d) What are the potential reasons for the difference between results reported in the above
two regressions.
(e) Find the state that has the largest number for the execution variable in 1993. (The
variable exec is total executions in 1991, 1992, and 1993.) How much bigger is this value
than the next highest value?
(f) Estimate the equation using first differencing, dropping Texas from the analysis. Com-
pute the usual and heteroskedasticity-robust standard errors. Now, what do you find?
What is going on?
1
ECA5103 Assignment 3
2. The number of young children is believed to have a negative impact on mother’s labor supply.
To analyze the impact of having more than two children (> 2) on mother’s labor force
participation, a researcher runs the following linear probability model
pi = γ0 + γ1edumi + β1kid3i + ui, i = 1, . . . , n (1)
where pi = 1 if mother i’s working hour is larger than 0 and pi = 0 otherwise, edumi is years
of schooling, kid3i is a dummy variable that is = 1 if i has more that two children and = 0 if
not.
(a) Based on the variables in the data angrist90.dta, what are additional variables that
should be included in the regression for the OLS estimator βˆ1 to be consistent for β1,
the true effect of having more children on mother’s labor supply.
(b) Generate kid3i from the variable kidcount and estimate equation (1) and (2), and com-
pare the βˆ from these two regressions and derive the sign of the correlation between
father’s education and the kid3i.
pi = γ0 + γ1edumi + γ2edufi + β2kid3i + ui, i = 1, . . . , n (2)
(c) Use the data to confirm your answer to the previous question.
(d) Is multi2nd a valid instrument for variable kid3? Explain your answer.
(e) Use variable multi2nd (=1 if 2nd birth was twins) as the IV to estimate equation (1)
and explain its meaning.
(f) Assume eduf is the only missing variable in equation (1), compare the biases of the IV
estimate with the biases of the OLS from equation (1).