A1842
Title: Bayesian analysis of Cox regression with partly linear covariate effects via reversible jump MCMC.
Authors: Chun Yin Lee - Hang Seng University of Hong Kong (Hong Kong) [presenting]
Kin Yau Wong - Hong Kong Polytechnic University (Hong Kong)
Hengtao Zhang - Guangdong Ocean University (China)
Yuanke Qu - Guangdong Ocean University (China)
Abstract: The partially linear Cox-type regression model provides a robust and flexible framework for exploring potential nonlinear effects of continuous variables on censored outcomes in the context of complex diseases. Most contemporary research studies estimation methods for such models based on the Frequentist paradigm, necessitating the selection of bandwidth and/or the number of spline basis functions for smoothing. A Bayesian estimation approach is proposed that eliminates this requirement by employing the reversible jump Markov chain Monte Carlo (RJMCMC) algorithm. The proposed method can inherently estimate both the number and location of knots in the unknown function in a data-adaptive manner during the posterior inference process, thereby enhancing applicability and predictive accuracy. Finite-sample performance of the proposed method is evaluated through simulation studies. The effectiveness of the proposed method is illustrated through analyses of two medical datasets.
