Robert M. Sauer and Christopher R. Taber, "Understanding Women's Wage
Growth using Indirect Inference with Importance Sampling," Journal of
Applied Econometrics, Vol. 36, No. 4, 2021, pp. 453-473.
All files are zipped in the file st-files.zip. ASCII text files are in
DOS format. Since there is also one .dta file, Unix/Linux users should
be careful when converting these.
The main data set is fordatw.raw. It can be found in the Data
directory. It is an ASCII files with variables separated by spaces. It
contains data on women from the Survey of Income and Program
Participation.
The variables in the file are:
-person id
-Potential Experience
-Dummy for currently employed
-years of education
-Dummy for currently marries
-Number of Children Less than 18
-Number of Children Less than 7
-log wage
-Total number of Children mom has given birth to
-Number of children greater than 18
-dummy variable for any children
-age of youngest child
-difference in age between youngest and oldest child
-dummy variable for having a baby since last wave
-wave number
This file, Table 1, and all the auxiliary parameters in our model can
be produced using the program fordat.do, which uses the stata file
femdat.dta. Figure 1 can be reproduced using mkfig1.do.
We have included all of the code to produce the numbers in our figures
and tables. These are produced with a combination of programs for
Stata, Fortran, and Julia. Each directory includes its own README
file in pdf versions which explain the files in those directories.
In general the code would be run in the following order of directories:
-- MonteCarlo produces Monte Carlo results.
-- Aux_moments constructs auxiliary moments and their
variance/covariance matrix to be used in estimation.
-- Estimate does the bulk of the work providing the code to estimate
the model and simulate the counterfactuals.
-- Tables contains the code to calculate standard errors of the
parameters and produce the tables.
-- Figures contains the code to produce the figures.