Analysis of Progressively Censored Competing Risks Data

Abstract

In several studies in Survival Analysis, the cause of failure/ death of items or individuals may be attributable to more than one cause. In this chapter, we consider the competing risks model when the data is progressively Type-II censored. We provide different techniques for the analysis of the model under the assumption of independent causes of failure and exponential lifetimes. The maximum likelihood estimators of the different parameters and the UMVUE's are obtained. In addition, the exact distributions of the different estimators are derived. We also derive the UMP and UMPU test for the equality of the failure rates of the competing risks.We consider the Bayesian estimation using the Inverse Gamma distribution as a prior. To assess the performance of all these estimators, confidence intervals are developed using the exact, asymptotic, and bootstrap distributions. In the Bayesian context, we develop credible intervals for the parameters. The different methods are compared through a simulation study, and the analysis of a real dataset. Finally, we also provide some insight into inference under the Weibull model and dependent causes of failure.

Inference for Weibull distribution under adaptive Type-I progressive hybrid censored competing risks data

“…We reanalyze a real data set analyzed by Hoel (1972), and reused by Kundu et al (2004). Also, a simulations study is used to compare the performance of the different estimators, different confidence intervals using different parameter values and different schemes.…”

Analysis of Generalized Exponential Distribution Under Adaptive Type-II Progressive Hybrid Censored Competing Risks Data

“…In this section, we analyze one data set which was originally analyzed by Hoel [6] and later by Kundu et al [7], Pareek et al [4], Cramer and Schmiedt [9], Hemmati and Khorram [11] and Ashour and Nassar [17]. The data was obtained from a laboratory experiment in which male mice received a radiation dose of 300 roentgens at 35 days to 42 days (5-6 weeks) of age.…”

Nonparametric predictive pairwise comparison with competing risks

“…A special topic that involves right-censored data is progressive censoring (Balakrishnan and Aggarwala, 2000;Kundu et al, 2004), where, during a lifetime experiment, non-failing units are withdrawn from the experiments. This could be done to save cost or time, but it may also be useful, at the moment a unit fails, to study the unit in detail in comparison with units in the same experiment that have not failed, to get better knowledge about the underlying cause of failure.…”