(a) Run the regression hrsempit = β0 + β1 grant it + β21( year = 1988) + β3Ei + uit where Ei is a dummy variable for being a treatment (i.e. someone who would receive the grant in 1988). (b) Run the fixed effect regression with firm fixed effects θi: hrsempit = θi + β1 grant it + β21( year = 1988) + uit (c) Construct the 4 means of controls and treatments, before and after, and es- timate the difference in difference with means. (d) Do you get exactly the same answer, why or why not? (e) Now include other controls to estimate the D-i-D regression. Justify what- ever you include and interpret. Provide line by line c
implement a D-i-D in this problem. Load the dataset on STATA: use http://www.stata.com/data/jwooldridge/eacsap/jtrain1 This has data on firms and the amount of job training they get. Only use the data from 1987 and 1988. Carefully study the data before you proceed. Construct the D-i-D estimator in different ways:
-
(a) Run the regression
hrsempit = β0 + β1 grant it + β21( year = 1988) + β3Ei + uitwhere Ei is a dummy variable for being a treatment (i.e. someone who would receive the grant in 1988).
-
(b) Run the fixed effect regression with firm fixed effects θi: hrsempit = θi + β1 grant it + β21( year = 1988) + uit
-
(c) Construct the 4 means of controls and treatments, before and after, and es- timate the difference in difference with means.
-
(d) Do you get exactly the same answer, why or why not?
-
(e) Now include other controls to estimate the D-i-D regression. Justify what- ever you include and interpret.
Provide line by line code for STATA and the solution
Trending now
This is a popular solution!
Step by step
Solved in 3 steps
