Search

In this paper, we present case deletion and mean shift outlie r models for linear mixed measure- ment error models using the corrected likelihood of Nakamur a (1990). We derive the corrected score test statistic for outliers detection based on mean sh ift outlier models. Furthermore, several case deletion diagnostics are constructed as a tool for influ ence diagnostics. It is found that they can be written in terms of studentized residuals of model, er ror contrast matrix and the inverse of the response variable covariance matrix. Our influence diag nostics are illustrated through a real data set., Peer Reviewed

This document studied optimally bounded score functions for estimating regression parameters in a generalized linear model. This work extends results obtained by Krasker & Welsch (1982) for the linear model and provides a simple proof of Krasker and Welsch's first order condition for strong optimality. The application of these results to logistic regression is studied in some detail with an example given comparing the bounded influence estimator with maximum likelihood. Additional keywords: Outliers; Robustness; Influentral points. (Author), Prepared in cooperation with Cornell Univ., Ithaca, NY. Dept. of Economic and Social Statistics.