首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Testing for heteroscedasticity of exponential correlation mixed-effects linear models based on M-estimation
Authors:Hui-Hui Sun
Institution:School of Mathematics and Science, Yancheng Teachers University, Yancheng, China
Abstract:Homogeneity of variance is a basic assumption in longitudinal data analysis. However, the assumption is not necessarily appropriate. In this paper, Fisher scoring method is applied to get M-estimator in the exponential correlation mixed-effects linear model. The score tests for heteroscedasticity and correlation coefficient based on M-estimator are then studied. Monte Carlo method is applied to investigate the properties of test statistics. At last, the methods and properties are illustrated by an actual data example.
Keywords:Exponential correlation  Heteroscedastic errors  Longitudinal data  M-estimation
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号