部分区间删失数据下比例风险模型的贝叶斯变量选择  

Bayesian variable selection for proportional hazards models with partially interval-censored data

作　　者：李纯净 赵昱榕 张淼 LI Chun-jing;ZHAO Yu-rong;ZHANG Miao(School of Mathematics and Statistics,Changchun University of Technology,Changchun 130012,China)

机构地区：[1]长春工业大学数学与统计学院,吉林长春130012

出　　处：《东北师大学报（自然科学版）》2023年第3期37-44,共8页Journal of Northeast Normal University(Natural Science Edition)

基　　金：国家自然科学基金青年基金资助项目(11901053);吉林省科技厅重大科技专项基金资助项目(20210301038GX);吉林省科技厅科技发展规划项目(20210101149).

摘　　要：通过结合spike and slab先验和非参数似然函数,提出了部分区间删失数据下比例风险模型的贝叶斯变量选择.引入泊松与0-1潜变量简化似然,进而运用EM算法求出了各个参数的最大后验估计.通过模拟和实证验证了该方法的有效性.By combining spike and slab priors and non-parametric likelihood functions,a Bayesian variable selection for proportional hazards models with partially interval-censored data is proposed.Poisson and 0-1 latent variables are introduced to simplify the likelihood,and then the EM algorithm is used to obtain the maximum a posteriori estimate of each parameter.The effectiveness of the method is verified by simulation and empirical.

关 键 词：部分区间删失 比例风险模型 潜变量 贝叶斯变量选择 EM算法 

分 类 号：O212[理学—概率论与数理统计]