Likelihood Inference for Step Stress Partially Accelerated Life Test Model with Type I Progressively Hybrid Censored Data from Generalized Exponential Distribution

Abstract

This thesis considers the statistical inference on the generalized exponential distribution parameters in presence of progressive Type-I censoring under partially accelerated life test. The maximum likelihood method is used to estimate the unknown parameters in the case of step-stress partially accelerated life tests. The performance of the estimators is investigated using simulation for certain simulation designs and sample sizes. The biases and mean square errors of the maximum-likelihood estimators are computed to assess the performance of the point estimators whereas coverage probability and expected length is used to assess the performance of the interval estimators. These interval estimators are derived using three classical approaches namely; asymptotic, Bootstrap percentile interval, and Bootstrap-t confidence intervals. Comparison between these three methods is also conducted. Further, reliability functions are derived for various time-t and the point and interval estimators are investigated. To illustrate the above, a data analysis is conducted. Finally, conclusions and ideas for possible future research are discussed in this thesis