Diagnostic measures for the restricted Liu estimator in linear measurement error models

Abstract

This article concentrated on the diagnostics measures to identify outlier observations using the restricted Liu estimator (RLE) of the vector of parameters in linear measurement error models (LMEMs). To this end, the mean shift outlier model (MSOM) has been determined to detect outliers and the corrected score test statistic. Also, using the RLE, generalized leverage matrices were constructed to identify high-leverage observations. Furthermore, to identify the influential observations, different statistics that are single-case deletion methods are used based on RLE. Finally, a simulation study and a real example are provided to illustrate the performance of the influence measures.