Burlig, Sudarshan, and Schlauch (2021) replication archive

This repository contains all data and code required to replicate Burlig, Sudarshan, and Schlauch (2021): "The impact of domestic travel bans on COVID-19 cases is nonlinear in their duration". The main text of the paper can be found in BSS_COVID.pdf, and the Appendix can be found in BSS_COVID_Appx.pdf.

Software

This repository uses both Stata and R. All code has been validated to run on Stata 16 MP and R 4.02, as of February 2021. We also utilize a variety of user-written packages for both programs, listed here:

• Stata:

  • gsort, missings, reclink, strgroup, rangestat, dm88_1, gr0075, gtools, and reghdfe. All of these files are available from ssc.
  • scheme-fb2.scheme is required for proper figure aesthetics. It is included in this repository. Users should install it into their ado folder.

• R:

  • sf, plyr, tidyverse, nngeo, rlist, rjson, rmapshaper, grDevices, rgdal, extrafont, stringi, exactextractr, raster, fasterize, maptools, and sp are all available from CRAN.
  • covoid is available from Github, using the following code snippet:

  # if necessary, install devtools:
  install.packages("devtools")
  # install covoid:
  devtools::install_github("cbdrh/covoid",build_vignettes = TRUE)]
  

• Inkscape:

  • Fig. 1 in the main text was partially assembled in Inkscape.

Code and data

This repository contains all code and data required to reproduce our estimates, figures, and tables in the main paper and Supplementary Information.

The full project, starting from raw data and ending with figures and tables, can be replicated by running MASTER_run_full_project.do. This calls all required .do files and .R scripts in order.

Replicator interested in starting from cleaned data and producing figures/tables only should run the MASTER_run_replication.do file, which fully replicates the results starting from the cleaned data. These two files call various subprograms, contained in the /Code folder.

Prior to running the MASTER file, researchers must set the indicated file paths at the top of this .do file. In addition, replicators must change the directory path (the setwd() step) at the top of each included .R script that they intend to run.

All data are stored in Box. Researchers should download and unzip this folder and assemble the replication folder structure as described below. The ZIP folder populates all datasets required for researchers to run MASTER_run_full_project.do and MASTER_run_replication.do.

MAIN PROJECT FOLDER
|-- Code
|   |-- Analyze
|   |-- Build
|   |-- Merge
|   |-- PO
|-- Data
|   |-- Raw
|       |--  Census
|            |--  India
|                 |--  D1
|                 |--  D2
|            |--  Indonesia
|            |--  Kenya
|            |--  Philippines
|            |--  South Africa
|       |--  Covid
|       |--  Misc
|       |--  Rasters
|                 |--  World pop
|       |--  Shapefiles
|            |--  Districts
|            |--  Non india
|                 |--  China
|                 |--  Indonesia
|                 |--  Kenya
|                      |--  adm1
|                      |--  adm2
|                      |--  county
|                 |--  Philippines
|                 |--  South Africa
|   |-- Generated
|       |--  Intermediate
|            |--  Census
|            |--  Covid
|            |--  Map
|            |--  Shapefiles
|                 |--  Districts
|                      |--  aggregated
|                      |--  combined
|                      |--  separated
|            |--  Temp
|            |--  World pop
|       |--  Final
|            |--  Regression inputs
|            |--  Results
|            |--  Shapefiles
|-- Outputs
|       |--  Figures
|       |--  Maps
|       |--  Tables

Contact

If you have remaining questions about the code described here, please contact Fiona Burlig.