Data & Code

Replication Data for Papers

Poverty accounting (with D. de Crombrugghe and A. Szirmai)

  • Poverty data in levels (dta, csv, xls)
  • Full replication set (zip), see below for fhetprob estimator.

Do weak institutions prolong crises? On the identification, characteristics, and duration of declines during economic slumps (with D. de Crombrugghe and A. Szirmai)

  • Full replication set coming soon.
  • Stata/Mata code for restricted structural break search (coming soon).

Fueling conflict? (De)Escalation and bilateral aid

  • Stata code for estimating dynamic ordered probit models with endogeneity and quasi fixed-effects/CRE (coming soon).

R Modules

Conley standard errors for massive data sets (on Github but still undocumented)

  • Parallelized Rcpp/C++ code for computing Conley errors. Can be used on panels and cross-sections. Works with lm() and lfe() objects. I use it on cross-sections of Kenyan census data with > 1 million observations.
  • It’s a fork from Darin Christensen and Thiemo Fetzer’s excellent package. I added the implicit parallelization with RcppParallel and generalized the program to add some flexibility.

Stata Modules

fhetprob: A fast QMLE Stata routine for fractional probit models with multiplicative heteroskedasticity

  • To install the software type the following at Stata’s dot prompt:
  • net install fhetprob, from(http://www.richard-bluhm.com/stata/)
  • To get the example data and pdf documentation (installs in current directory):
  • net get fhetprob, from(http://www.richard-bluhm.com/stata/)
  • Then type help fhetprob or read fhetprob.pdf for usage instructions.