Random effect models for repeated measures of zero-inflated count data

By Yongyi Min, Alan Agresti. 2005

400 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Statistics and Probability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Statistics and Probability

Citations in the corpus

A Bayesian model for repeated measures zero-inflated count data with application to outpatient psychiatric service use (2010, methodology): “However, as Min and Agresti (2005) note, the ZIP model can be unstable when zero deflation exists at a particular level of a categorical predictor, even if the data on the whole are zero inflated.”
Exposed: Venture Capital, Competitor Ties, and Entrepreneurial Innovation (2014, background): “…little within-firm variation and to zero-inflated Poisson models being unstable under fixed-effects specifications for clustered data like ours (see Min & Agresti (2005), who recommended random-effects specifications for ZIP models). package in R, which allowed for the estimation of random effects…”
Life history and behavioral type in the highly social cichlid Neolamprologus pulcher (2010, background): “How to deal with zero inﬂation in longitudinal data analysis is still debated (e.g., Martin et al. 2005; Min and Agresti 2005).”
On the Use of Zero-Inflated and Hurdle Models for Modeling Vaccine Adverse Event Count Data (2006, background): “Min and Agresti ( 2005 ) fit several random effect hurdle models to pharmaceutical data.”
Foundations of Linear and Generalized Linear Models (2015, methodology): “5 Hurdle model, and ZIP versus ZINB: Mullahy (1986) proposed the hurdle model using the truncated Poisson or geometric distribution. Ridout et al. (2001) provided a score test of the ZIP model against the ZINB alternative.”
Two‐part regression models for longitudinal zero‐inflated count data (2010, methodology): “According to Min & Agresti (2005) and Greene (2009), we allow for an endogenous decision step by including correlated random effects in the model specification.”