EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: On goodness-of-fit test for beta-binomial regression models Authors:  Junghwan Kim - Institute of Water Resources System, Inha University (Korea, South) [presenting]
Woojoo Lee - Inha University (Korea, South)
Abstract: Beta-binomial regression models have been widely used for analyzing over-dispersed binomial data. However, they are not always flawless models that can capturing all types of extra-binomial variation. Like other regression models, statistical inferential procedures from beta-binomial regression models can be problematic if an important co-variate is not included or some heterogeneity is shown in individuals. To test the existence of the hidden heterogeneity in beta-binomial regression models, we provide score test statistics and investigate their performance in terms of type I error and power.