Dynamic GMM for Gravity Model

[Iman Pal]

I am trying to run dynamic GMM and find Arellano–Bond estimator for Gravity model. I am currently using Stata 12.0. I am using a set of 24 countries over a period of 24 years (1995-2018). The columns in my excel sheet are as follows: reporter country, partner country, import, export, NX, reporter GDP, Partner GDP, Reporter population, partner population, distance, real ER, and a set of dummy variables (common language, common religion, FTA, border, colony, development, currency). Please tell me what commands to use.

[Sebastian Kripfganz]

24 countries over 24 years does not seem to be a data set that is particularly suitable for the Arellano-Bond estimator. You would typically need to have large N (number of countries) relative to small T (number of time periods) for this estimator to provide reliable results.

I recommend to use a simple IV estimator with ivregress (or ivreg2) instead.

Nevertheless, the following presentation provides more information on dynamic panel model GMM estimation in Stata:

• Kripfganz, S. (2019). Generalized method of moments estimation of linear dynamic panel data models. Proceedings of the 2019 London Stata Conference.