Testing for Heteroscedasticity in Nonlinear Models of Longitudinal Data

摘要：

In longitudinal data analysis, homogeneity of variance is a basic assumption in the model random errors. However, this assumption is not necessarily appropriate. Zhang & Weiss (2000), Lin & Wei (2003) tested for heteroscedasticity in models with random effects based on longitudinal data. Li Yong et al. (2007) discussed the testing for heteroscedasticity in two phased linear regression models. This paper extends their results to nonlinear models of longitudinal data. From studying composite and individual tests in or between populations, we get the score statistics and their adjustment forms of tests. The plasma concentrations data (Davidian & Giltian, 1995) is used to illustrate the applying of our testing methods.

关键词： Longitudinal data nonlinear model heteroscedasticity score test.

作者: CAO Chun-zheng ZHU Xiao-xin MEN Ke-pei

作者单位: College of Math & Physics, Nanjing University of Information Science & Technology, Nanjing, 210044, China

会议类型: 国际会议

会议名称: 2008年国际应用统计学术研讨会(2008 International Institute of Applied Statistics Studies)

会议地点: 烟台

会议语种:英文

页码: 1-4

在线出版日期: 2008-08-14（万方平台首次上网日期，不代表论文的发表时间）