Robust estimation of zero-inflated count models

Kevin E. Staub; Rainer Winkelmann

Applications of zero-inflated count data models have proliferated in empirical economic research. There is a downside to this development, as zero-inflated Poisson or zero-inflated Negative Binomial Maximum Likelihood estimators are not robust to misspecification. In contrast, simple Poisson regression provides consistent parameter estimates even in the presence of excess zeros. The advantages of the Poisson approach are illustrated in a series of Monte Carlo simulations. -- excess zeros ; Poisson ; overdispersion ; Negative Binomial regression

MoreLess

Year of publication:	2009
Authors:	Staub, Kevin E. ; Winkelmann, Rainer
Publisher:	Zürich : Univ., Sozialökonomisches Inst.
Subject:	Generalisiertes lineares Modell \| Generalized linear model \| Stochastischer Prozess \| Stochastic process \| Maximum-Likelihood-Schätzung \| Maximum likelihood estimation \| Monte-Carlo-Simulation \| Monte Carlo simulation \| Robustes Verfahren \| Robust statistics \| Theorie \| Theory