Tutorial 7: Stata Application of Advanced Panel Data Method The University of Queensland
7360
Instructor: Rigissa Megalokonomou
Problem I: The Deterrent Effects of Execution on Murder
Use the data in murder.dta for the following questions. This includes the state level data on murder rates and executions.
(i) Consider the unobserved effects model
mrdrteit = thetat + ¦Â1execit + ¦Â2unemit + ci + uit (1)
where thetat simply denotes different year intercepts and ci is the unobserved state effect. If past executions of convicted murderers have a deterrent effect, what should be the sign of ¦Â1? What sign do you think ¦Â2 should have?
(ii) Using just the years 1990 and 1993, estimate the equation from part (i) by pooled OLS. Ignore the serial correlation problem in the composite errors. Do you find any evidence for a deterrent effect?
(iii) Now, using 1990 and 1993, estimate the equation (3) by fixed effects. Now, is there any evidence of deterrent effect?
(iv) Now, using 1990 and 1993, estimate the equation (3) by first differencing. Use the heteroskedasticity- robust standard error and do not include a constant. Is there any evidence of deterrent effect?
(v) Find the state that has the largest number for the execution variable in 1993. How much bigger is this value than the next highest value?
(vi) Estimate the equation using first differencing, while dropping Texas from the analysis. Use the usual and heteroskedasticity-robust standard errors. Now, what do you find?
(vii) Use all three years of data and estimate the model by fixed effects. Include Texas in the analysis. Problem II: A Study of the effect of small class with Tennessee
Project STAR
Background The data set for this exercise comes from the paper by Krueger (1999) ¡±Experimental Es- timates of Education Production Functions¡±, published in the Quarterly Journal of Economics, vol. 114, No.2, p. 497-532. He analyzes data on 11,600 students and their teachers who were randomly assigned to different size classes from kindergarten through 3rd grade. Use the data in starwide.dta for the following
1
questions. The data include students from kindergarten to 3rd grade.
(i) The loaded dataset is in a wide form. Construct two variables that sum up years in small class as well as sum up years in aide class.
(ii) Convert wide data to long data form with stdntid as cross-section and grade as time.
(iii) Now we have four year data for each students. Define data as panel data with stdntid as cross- section and grade as time.
(iv) Replicate the following figures from Krueger (1999). In particular, construct the four densities for regular and small classes in each grade you have in the dataset. ( the dummy ¡±regular¡± indicates if a student is assigned to a regular class and ¡±small¡± indicates if a student is assigned to a small class )
2