Abstract
Excessive zeros are common in practice and may cause overdispersion and invalidate inference when fitting Poisson regression models. There is a large body of literature on zero-inflated Poisson models. However, methods for testing whether there are excessive zeros are less well developed. The Vuong test comparing a Poisson and a zero-inflated Poisson model is commonly applied in practice. However, the type I error of the test often deviates seriously from the nominal level, rendering serious doubts on the validity of the test in such applications. In this paper, we develop a new approach for testing inflated zeros under the Poisson model. Unlike the Vuong test for inflated zeros, our method does not require a zero-inflated Poisson model to perform the test. Simulation studies show that when compared with the Vuong test our approach not only better at controlling type I error rate, but also yield more power.
Original language | English (US) |
---|---|
Pages (from-to) | 1157-1169 |
Number of pages | 13 |
Journal | Statistical Methods in Medical Research |
Volume | 28 |
Issue number | 4 |
DOIs | |
State | Published - Apr 1 2019 |
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by NIH grants R33 DA027521, R01GM108337, and P20GM109036.
Keywords
- Vuong test
- Zero-inflated Poisson (ZIP)
- hypothesis test
- power
- type I error
ASJC Scopus subject areas
- Health Information Management
- Epidemiology
- Statistics and Probability