Local Influence Diagnostics of Replicated Data with Measurement Errors

Abstract

Replicated data with measurement errors frequently exist in various scientific fields. In this work, we propose a replicated measurement error model for such data under scale mixtures of normal distributions. We consider local influence diagnostics to detect and classify outliers in the data through different perturbation schemes. \u00a0A \u00a0simulation \u00a0study \u00a0and \u00a0an \u00a0application \u00a0confirm \u00a0the \u00a0effectiveness \u00a0and \u00a0robustness \u00a0of \u00a0the \u00a0diagnostic method.