A Bayesian Nonparametric Estimator of Survival Probability Assuming Increasing Failure Rate

摘要

Bayesian nonparametric estimators of the survival function, the failure rate function, and the density function are obtained using jump processes as prior distributions on the space of increasing failure rate functions. The jump processes are more appealing intuitively than previously used Dirichlet processes, integral gamma process, and processes neutral to the right and have a meaningful physical interpretation. Examples are given and the estimates are compared with the maximum likelihood estimates. In addition, the Bayesian nonparametric estimators are presented for arbitrarily right-censored observations. (Author)