Mohitosh Kejriwal, Xiaoxiao Li, and Evan Totty, "Multidimensional
Skills and the Returns to Schooling: Evidence from an Interactive
Fixed Effects Approach and a Linked Survey-Administrative Dataset,"
Journal of Applied Econometrics, Vol. 35, No. 5, 2020, pp. 548-566.

The file klt-files.zip contains three folders: "programs", "mydata",
and "output". All files are ASCII files in DOS format. Unix/Linux
users should us "unzip -a".

-- "programs" contains all of the Stata and MATLAB programs needed to
replicate the results in the paper. There is a master file named
"master.do" that calls all of the other programs. The code can be run
all at once by running master.do, or piece-by-piece by running the
"p#_[...]" programs sequentially. Note that there is a sub-folder
named "matlabprograms" that contains several matlab function files
called by the main MATLAB programs.

-- "mydata" stores intermediate data files used by the Stata and
MATLAB programs. It also includes two publicly available data files
used in the code described below. 

-- "output" stores the results from the paper. After running all of
the programs, this folder should contain the following files:

  1. Table1.csv - results from Table 1 in the paper
  2. Table3.csv - results from Table 3
  3. Table4.csv - results from Table 4
  4. Table5.csv - results from Table 5
  5. Table6.xls - results from Table 6
  6. Table6_pvalues.xls - p-values associated with Table 6
  7. Table7.xls - results from Table 7
  8. TableD1.csv - results from online appendix Table D1
  9. TableD2.csv - results from online appendix Table D2
 10. TableD3.csv - results from online appendix Table D3
 11. TableD4.csv - results from online appendix Table D4
 12. TableCDstats.csv - CD tests statistics for Tables 3, D1, and D4
 13. Figure1.gph - Figure 1 in the paper
1 4. Figure2.gph - Figure 2 in the paper


The data used in this paper are confidential U.S. Census Bureau data
and thus cannot be provided on the Journal of Applied Econometrics
website. However, the Census Bureau does provide free and relatively
easy access to this data that does not require access to a Federal
Statistical Research Data Center. The data used in the paper are
linked SIPP-IRS-SSA data, known as the "Gold Standard File" (GSF). 
The Census Bureau provides a synthetic version of the GSF data to the
public for free, known as SIPP Synthetic Beta (SSB). Users must apply
for access, but access is typically granted within 1-2 days. The
synthetic file has the same variable names and number of records as
the GSF. External users may use these data to build their code, and 
then may submit the code to the Census Bureau for "validation," in
which the code is run on the internal data and the results are
released back to the user. More information on the GSF and SSB can be
found at

https://www.census.gov/programs-surveys/sipp/guidance/sipp-synthetic-beta-data-product.html

and in "The Creation and Use of SIPP Synthetic Beta v7.0" (Bendetto,
Stanley, and Totty; 2018).

Our analysis also uses two publicly available files for information on
minimum wages and inflation. The minimum wage data are in the
"VZ_state_annual" file from "Historical State and Sub-State Minimum
Wage Data" (Vaghul and Zipperer, 2016):

https://www.equitablegrowth.org/working-papers/historical-state-and-sub-state-minimum-wage-data/.

The inflation data are in the non-seasonally-adjusted CPI-U-RS file
from the Bureau of Labor Statistics (the first 6 rows were deleted so
the file could be read more easily by Stata):

https://www.bls.gov/cpi/research-series/home.htm.

These two datasets, along with the code to re-produce the results in
the paper, can be found in klt-files.zip. The two datasets are
provided as .csv files and the code includes a section that imports
them into Stata .dta files.

Benedetto, G., J.C. Stanley, and E. Totty (2018). The Creation and Use
of the SIPP Synthetic Beta v7.0. CES Technical Notes Series 18-03,
Center for Economic Studies, U.S. Census Bureau.


Evan Totty
evan.scott.totty [AT] census.gov