Optimum Bayesian Estimators for Proposed Nonparametric Proportional Hazards Model

Abstract

Through investigation and research in the literature of survival models generally; and proportional hazards models in particular, It turns out that there has never been a study of Nonparametric proportional hazards models, as well as a lack of adopted optimization theory through the application Bayesian optimization criteria in determining the optimal design that gives the estimators of the parameters of nonlinear models (proportional hazards models) with the least variance, it has not been studied previously Using this theory to estimate the parameters of survival models in Iraq or other Arab countries until now as we know. So we saw adopt the criteria of optimization Bayesian (DB, CB, AB) in reaching an optimal design to estimate the parameters of the Proposed nonparametric proportional hazards models of patients with myocardial infarction, which is among the most serious diseases that threaten human life by using the Kaplan Meier estimator of hazards function to represent the baseline hazards function .The study found that the effect of hazards factors represented by (Triglycerides (Trig), Low Density Lipoprotein (VLDL) and Low Blood Pressure Examination) adverse on the survival time of patients with myocardial infarction through the parameters of the Proposed nonparametric model estimated from the optimal designs obtained according to all the criteria of optimization studied this match with the medical reality.