Bayesian Analysis of the Generalized Additive Proportional Hazards Model: Asymptotic Studies

In this paper, we study Bayesian asymptotic properties of the proportional hazards model where link function is modeled by generalized additive model. As standard is, a useful tool in finding nonlinearity covariate effects to survival times. We develop data-dependent sieve prior for and use bootstrap baseline hazard function. prove that posterior contraction rate minimax optimal up logn term when carefully selected. By analyzing simulated as well real data, verify our theoretical results compare with exisiting algorithms illustrate proposed inference tool.