ppmlhdfe: Fast Poisson Estimation with High-Dimensional Fixed Effects

TL;DR

The paper introduces ppmlhdfe, a Stata command that enables fast and robust estimation of Poisson regression models with multiple high-dimensional fixed effects, leveraging optimized algorithms and techniques.

Contribution

It presents a new implementation of Poisson regression with high-dimensional fixed effects that is faster and more robust than existing methods.

Findings

Significantly faster estimation times for high-dimensional fixed effects.

Enhanced robustness in maximum likelihood estimation checks.

Compatibility with reghdfe for similar syntax and functionalities.

Abstract

In this paper we present ppmlhdfe, a new Stata command for estimation of (pseudo) Poisson regression models with multiple high-dimensional fixed effects (HDFE). Estimation is implemented using a modified version of the iteratively reweighted least-squares (IRLS) algorithm that allows for fast estimation in the presence of HDFE. Because the code is built around the reghdfe package, it has similar syntax, supports many of the same functionalities, and benefits from reghdfe's fast convergence properties for computing high-dimensional least squares problems. Performance is further enhanced by some new techniques we introduce for accelerating HDFE-IRLS estimation specifically. ppmlhdfe also implements a novel and more robust approach to check for the existence of (pseudo) maximum likelihood estimates.