Skip to content

Latest commit

 

History

History
76 lines (58 loc) · 5.21 KB

README.md

File metadata and controls

76 lines (58 loc) · 5.21 KB
Raw

Volatility-and-Development-replication

Replication package for

Koren, Miklós, and Silvana Tenreyro, 2007. “Volatility and Development.” Quarterly Journal of Economics. 122(1), pp. 243-287.

Please cite the above paper when using this code or data. All code (*.do) is licensed under the 3-clause BSD license. Derived data (data/derived/) are licensed under the Open Database License 1.0. Please respect the original licenses of raw datasets, provided here for your convenience, as specified below.

Data Availability and Provenance Statements

This research is based on publicly available data, reproduced in this repository.

INDSTAT2 was downloaded from UNIDO (UNIDO 2002). Central European University has purchased "secondary dissemination" license, which permits us to share large portions of the data for replication purposes.

World Development Indicators (World Bank 2003) is available with a Creative Commons License.

Penn World Table 6.1 (Heston, Summers and Aten 2002) does not provide an explicit license, but more recent versions are licensed under Creative Commons.

Dataset list

Data file Source Notes Provided
data/unido/*.dta UNIDO (2002) Converted to Stata format Yes
data/wdi/wdinew.dta World Bank (2003) Converted to Stata format Yes
data/pwt/*.dta Heston et al (2002) Converted to Stata format Yes
data/derived/*.dta Own calculations Yes

Computational Requirements

Software Requirements

  • Stata (last run with version 16.0)
  • GNU Make (optional)

Memory and Runtime Requirements

The code was lust run on a 4-core Mac with 8GB RAM. Computation took a few minutes.

Description of programs

The Makefile shows the order in which programs should be run and the dependency among them.

  • The program meanreturns.do loads all the data and creates sector-country-year level shares and productivity shocks. These are saved in the folder data/derived/.
  • The program preparefactors.do extracts the latent factors from productivity shocks. These are saved under data/derived/allfactors.dta.
  • The program specmeasures.do computes the risk measures as explained in the article. These are saved in data/derived/specmeasures.dta.
  • The program createallgraphs.do creates all panels in Figure IIa, IIb and III of the article. These are saved in graphs/ in .eps format.
  • The program drawgraphs.do is an auxiliary program to create lowess figures with logged dependent variables. It is called by createallgraphs.do.

Instructions to Replicators

If you have GNU Make on your system, type make in the folder containing the Makefile. This assumes that yout Stata can be called from the shell by typing stata. If not, edit the Makefile accordingly.

If you do not have GNU Make, call the .do files directly from Stata as follows.

cd <path to replication folder>
do meanreturns.do
do preparefactors.do
do specmeasures.do
do createallgraphs.do

List of Tables and Figures

Figure/Table # Program Line Number Output file Note
Figure IIa (top left) createallgraphs.do 28 lowess_Sectoral_Risk_loggdp.eps
Figure IIa (top right) createallgraphs.do 28 within_Sectoral_Risk_loggdp.eps
Figure IIa (bottom left) createallgraphs.do 28 lowess_Idiosyncratic_Risk_loggdp.eps
Figure IIa (bottom right) createallgraphs.do 28 within_Idiosyncratic_Risk_loggdp.eps
Figure IIb (top left) createallgraphs.do 28 lowess_Sector_Country_Covariance_loggdp.eps
Figure IIb (top right) createallgraphs.do 28 within_Sector_Country_Covariance_loggdp.eps
Figure IIb (bottom) createallgraphs.do 38 lowess_Country_Risk_loggdp.eps
Figure III (top left) createallgraphs.do 28 lowess_Herfindahl_loggdp.eps
Figure III (top right) createallgraphs.do 28 within_Herfindahl_loggdp.eps
Figure III (bottom left) createallgraphs.do 28 lowess_Average_Variance_loggdp.eps
Figure III (bottom right) createallgraphs.do 28 within_Average_Variance_loggdp.eps

References