Datasets
Export Similarity Index: Bilateral (country-country) measure of similarity of export baskets, on a yearly basis. Download here. If used, please cite the following paper:
Bahar, D., R. Hausmann, and C.A. Hidalgo. "Neighbors and the Evolution of the Comparative Advantage of Nations: Evidence of International Knowledge Diffusion?" Journal of International Economics, Volume 92, Issue 1, January 2014, pp. 111-123.
Industry-level Knowledge Intensity Measures: Combines O*NET worker-level information to quantify the tacit knowledge in each industry, for SITC and NAICS (4-digit). Download here. If used please cite the following two papers:
Bahar, D. "The hardships of long distance relationships: time zone proximity and the location of MNC’s knowledge-intensive activities". Journal of International Economics
Bahar, D. "Measuring knowledge intensity in manufacturing industries: a new approach". Applied Economic Letters. Volume 26, Issue 3, 2019, pp. 187-190.
Scripts
IVREG2HDFE: Stata module to estimate an Instrumental Variable Linear Regression Model with two High Dimensional Fixed Effects
[Download here or install in STATA by typing ssc install ivreg2hdfe]
Abstract: This command builds on the command reg2hdfe and ivreg2 for estimation of a linear instrumental variables regression model with two high dimensional fixed effects. The command is particulary useful when an instrumental variable approach is required in particularly large datasets, because it removes the high dimensional fixed effects from the data in the first step. Then it runs the IV model with the transformed variables after the proper demeaning.
REGWLS: Stata module to estimate Weighted Least Squares with factor variables
[Download here or install in STATA by typing ssc install regwls]
Abstract: This command incorporates support for factor variables, extending the command wls0 (Ender, UCLA). It also allows for the absorption of one fixed effects using the algorithm of the command areg. This is particularly useful when in the need of running a Weighted-Least Squares (WLS) model that requires a large number of dummy variables.