Local Influence in Regression Models with Measurement Errors and Censored Data Considering the Student-t Distribution

In this paper, the local influence approach is studied in regression models with measurement errors for multivariate censored responses under the Student-t distribution. The multivariate Student-t distribution and the multivariate normal, distributions of the independent normal class, are studied and used to compare various measuring instruments. The ECM algorithm is used to obtain maximum likelihood estimates of the model parameters and using the log-likelihood function of the complete data we obtain measures of local influence based on the methodology proposed by Zhu and Lee (Journal of the Royal Statistical Society, Series B 63:121-126, 2001) and Lee and Xu (Computational Statistics and Data Analysis 45:321-341, 2004). Finally, the described methodologies are used in real data analysis that illustrates the usefulness of the approach.