首页 | 本学科首页   官方微博 | 高级检索  
     


Influence diagnostics for the structural errors-in-variables model under the Student-t distribution
Authors:Manuel Galea   Heleno Bolfarine  Filidor Vilcalabra
Abstract:The influence of observations on the parameter estimates for the simple structural errors-in-variables model with no equation error, under the Student-t distribution, is investigated using the local influence approach. The main conclusion is that the Student-t model with small degrees of freedom is able to incorporate possible outliers and influential observations in the data. The likelihood displacement approach is useful for outlier detection, especially when a masking phenomenon is present and the degrees of freedom parameter is large. The diagnostics are illustrated with two examples.
Keywords:
本文献已被 InformaWorld 等数据库收录!
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