Modeling Zero-Inflated Count Data with Underdispersion and Overdispersion
نویسنده
چکیده
A common problem in modeling count data is underdispersion or overdispersion. This paper discusses the distinction between overdispersion due to excess zeros and overdispersion due to values that are greater than 0. It shows how to use exploratory data analysis to determine the dispersion patterns and that the dispersion patterns can change depending on the predictors and the subpopulation that are included in the analysis. Further, the paper discusses how to fit zero-inflated models using PROC NLMIXED and compares the model fit. The data is from the National Health and Nutrition Examination Survey (NHANES 2003-2004).
منابع مشابه
Spatial modeling of data with excessive zeros applied to reindeer pellet‐group counts
We analyze a real data set pertaining to reindeer fecal pellet-group counts obtained from a survey conducted in a forest area in northern Sweden. In the data set, over 70% of counts are zeros, and there is high spatial correlation. We use conditionally autoregressive random effects for modeling of spatial correlation in a Poisson generalized linear mixed model (GLMM), quasi-Poisson hierarchical...
متن کاملZero-inflated Poisson regression mixture model
Excess zeros and overdispersion are commonly encountered phenomena that limit the use of traditional Poisson regression models for modeling count data. The focus of this paper is on modeling count data in the case that a population has excess zero counts and also consists of several sub-populations in the non-zero counts. The proposed zero-inflated Poisson regression mixture model accounts for ...
متن کاملScore tests for zero-inflation and overdispersion in two-level count data
In a Poisson regression model, where observations are either clustered or represented by repeated measurements of counts, the number of observed zero counts is sometimes greater than the expected frequency by the Poisson distribution and the non-zero part of count data may be overdispersed. The zero-inflated negative binomial (ZINB) mixed regression model is suggested to analyze such data. Prev...
متن کاملA Comparison of Two Test Statistics for Poisson Overdispersion/Underdispersion
Within the family of zero-inflated Poisson distributions, the data has Poisson distribution if any only if the mean equals the variance. In this paper we compare two closely related test statistics constructed based on this idea. Our results show that although these two tests are asymptotically equivalent under the null hypothesis and are equally efficient, one test is always more efficient tha...
متن کاملModeling and Simulation of Count Data
Count data, or number of events per time interval, are discrete data arising from repeated time to event observations. Their mean count, or piecewise constant event rate, can be evaluated by discrete probability distributions from the Poisson model family. Clinical trial data characterization often involves population count analysis. This tutorial presents the basics and diagnostics of count mo...
متن کامل