基于Pena距离的双重广义线性模型的统计诊断

本文主要研究双重广义线性模型,考虑基于数据删除模型的参数估计和统计诊断,比较删除模型与未删除模型相应的诊断统计量之间的变化.首次提出基于双重广义线性模型下的Pena距离.通过一些模拟研究以及实例分析,比较不同诊断统计量判别异常点或强影响点的差异,研究结果表明本文提出的理论和方法是行之有效的.

Statistical Diagnostics for Double Generalized Linear Model Based on the Pena Distance

This paper mainly studies the double generalized linear model, considers the parameters estimation and statistical diagnosis based on the data deletion model, and compares the changes between the corresponding diagnostic statistic of the deleted model and the non-deleted model. F or the rst time, the Pena distance based on the double generalized linear model is proposed. Through some Monte Carlo simulation studies and a real data analysis, the di erences in di erent diagnosis statistics for discriminating between abnormal or strong in uence points were compared. The results show the proposed theories and methods in this paper are e ective.