Influential observations in the functional measurement error model

Abstract

In this work we propose Bayesian measures to quantify the influence of observationson the structural parameters of the simple measurement error model (MEM). Different influencemeasures, like those based onq-divergence between posterior distributions and Bayes risk, are studiedto evaluate the influence. A strategy based on the perturbation function and MCMC samples is usedto compute these measures. The samples from the posterior distributions are obtained by using theMetropolis–Hastings algorithm and assuming specific proper prior distributions. The results areillustrated with an application to a real example modeled with MEM in the literature.