Jackknife and Transformed Jackknife Empirical Likelihood Inferences for the Lifetime Performance Index with Missing Data
摘要
Abstract
An important topic in manufacturing industries is assessing the lifetime performance using a taken sample. But many factors may be implied to encounter with missing data. We propose the jackknife empirical likelihood (JEL) and transformed jackknife empirical likelihood (TJEL) methods to construct confidence intervals for the lifetime performance index with missing data. After using hot deck imputation, we apply the JEL and TJEL. Simulation studies are utilized to evaluate the proposed methods and some competitors in terms of coverage probability and average lengths of confidence intervals. A real data set is used to illustrate the proposed JEL and TJEL and the competitor methods.