Statistical Analysis For Exponential Distribution From Step-Stress Life Test Based On Uncomplete Data

The step-stress ALT (accelerated life test) is one of the most important method in Reliability Statistics. Getting the exact life data is a key step in the whole life test. But in practice, it is often impossiable continuously to inspect the testing process, we might only be able to obtain less information during the ALT. So it is necessary to study the inference when the information of data is missing.In the paper, we studies statistic analysis of grouped and censored data and missing data from step-stress life test. In chapter I, some inferences from such kinds of data presented by many authors are introduced, then we give the problems di-cussed in the paper. Chapter II presents the maximum likelihood estimator (MLE) and asymptotic confidence intervals estimation for exponential distribution under grouped and censored data when the accelerated life equation has p unknown parameters. Chapter III presents the optimum test plan for k-step test when the accelerated life equation has k unknown parameters under grouped and censored data. Inc chapter IV, we present the estimation of the product's mean life when the accelerated life equation has 2 unknown parameters under missing data. Examples are respectively given to illustrate the proposed methods in the paper.