Introduction

Count Data Models

The COUNTREG procedure provides regression models in which the dependent variable takes nonnegative integer count values. The COUNTREG procedure supports the following features:

  • Poisson regression

  • Conway-Maxwell-Poisson regression

  • negative binomial regression with quadratic and linear variance functions

  • zero-inflated Poisson (ZIP) regression

  • zero-inflated Conway-Maxwell-Poisson regression

  • zero-inflated negative binomial (ZINB) regression

  • fixed- and random-effects Poisson panel data models

  • fixed- and random-effects NB (negative binomial) panel data models

  • variable selection

  • Bayesian estimation and inference, including diagnostic plots

Last updated: June 19, 2025