Inferential Estimates from the One-parameter Half-normal Model

Abstract

In this paper, some inferential estimates are determined from a doubly censored sample, which follow the half normal model. A simulation study is conducted to illustrate the predictive results for a future response. A real life survival data study of 25 patients who were diagnosed with leukemia at a local hospital is utilized to illustrate the predictive results. The survival days of 19 patients with leukemia are used only for the statistical analysis. Six patients out of 25 patients were lost to follow-up, and they are the first three and the last three patients in the recorded sample. The survival days of the patients were recorded in order. The Bayesian sensitivity of the predictive survival days with respect to several values of the hyperparameter are reported. The summary results of the posterior parameter and hyperparameters are given by using of the Markov Chain Monte Carlo method.