Competing Causes of Failure for Weibull Lifetimes Under Type II Progressive Censoring

Abstract

<p>In life-testing experiments, specimens may fail due to more than one cause. Usually, the researcher who is interested in the primary cause of failure, treats the other causes of failure as censored data. In recent times, however, analysis is being done in the presence of competing causes of failure. Also, censoring becomes inevitable since it saves time and resources. In this project, we discuss inference for competing risks under Type II progressive censoring for Weibull lifetimes. We have considered a simple model where we assume equal shape parameters.</p> <p>This report is organized into six chapters. In Chapter 1, the concepts of progressive censoring and competing risks are introduced. The model is presented and then described. Notations used throughout the project are also given. Maximum likelihood equations are derived in Chapter 2 and the algorithm for solving them is also presented. Different confidence intervals for the parameters are discussed in Chapter 3 and the simulation results comparing the probability coverages and the lengths of the different confidence intervals are provided in Chapter 4. The performance of these different intervals is illustrated using a real dataset in Chapter 5. Finally, in Chapter 6, we provide a summary of the results and some extensions that may be carried out in the future.</p>