Guy Laroque and Bernard Salanié, "Identifying the Response of
Fertility to Financial Incentives", Journal of Applied Econometrics,
Vol. 29, No. 2, 2014, pp. 314-332.

The file LS-appendix.pdf contains a web appendix.

All data and programs are in R. See http://cran.r-project.org/. The
estimation and simulations programs rely on R packages Rmpi and Snow
to distribute the computing burden over available processors (on a
single machine or on a cluster.)

The data are in the R data file LaroqueSalanie.RData. For convenience,
we also include an ASCII file, LaroqueSalanie.txt. This contains the
same data as the R data file, with the names of the variables on the
first line, with the exception of the ten simulated values of the
incomes r10 and r11, which are in files r10.txt and r11.txt. The file
dictionary.txt gives a brief description of the variables.

mainProgram.R reads the data and calls mpiRoutines.R, which performs
the estimation and simulations.  The file auxilRoutines.R contains
some auxiliary routines, including three that manage some of the
interface with Rmpi and Snow.

There are two levels of numerical integration in mpiRoutines.R. The
first one relates to wages not being observed for non-employees, so
that incomes when working have to be simulated. We use for this
purpose the matrices r10 and r11, which collect 10 draws from the
pre-estimated distribution. The second level of integration is used to
compute the probability of fertility. We use m=50 quantiles from
N(0,1).
 
The initial values for the parameters are in file initialVals.RData; the
results are stored in estResults.RData.

All the RData files are zipped in the file ls-rdata.zip. All text
files are zipped in the file ls-code-data.zip. Text files are in Unix
format. Windows users may have to convert them to DOS format for use
with some programs.