序约束下单向分类方差分析模型的Bayes变量选择

吉林大学学报(理学版) ›› 2021, Vol. 59 ›› Issue (5): 1093-1100.

序约束下单向分类方差分析模型的Bayes变量选择

史海芳¹, 李聪², 姬永刚¹

1. 中国民航大学理学院, 天津 300300; 2. 吉林大学数学学院, 长春 130012

收稿日期:2021-01-12 出版日期:2021-09-26 发布日期:2021-09-26
通讯作者: 李聪 E-mail:li_cong@jlu.edu.cn

Bayesian Variable Selection in One-Way Classified ANOVA Models under Order Constraints

SHI Haifang¹, LI Cong², JI Yonggang¹

1. School of Science, Civil Aviation University of China, Tianjin 300300, China;
2. College of Mathematics, Jilin University, Changchun 130012, China

Received:2021-01-12 Online:2021-09-26 Published:2021-09-26

摘要/Abstract

摘要： 考虑序约束下单向分类方差分析模型的变量选择问题, 提出两种基于SSVS的Bayes变量选择方法, 并设计一种简单且易操作的Gibbs抽样算法进行后验抽样. 数值模拟和应用实例结果表明, 该方法效果较好.

关键词: Bayes变量选择, 序约束, 随机搜索变量选择

Abstract: We considered variable selection problem in one-way classified ANOVA under order constraints. We proposed two Bayesian variable selection method based on the stochastic search variable selection (SSVS), and designed a simple and easy to operate Gibbs sampling algorithm for a posteriors sampling. The results of numerical simulation and application examples show that the proposed method is effective.

Key words: Bayesian variable selection, order constraint, stochastic search variable selection

中图分类号:

O212.1

史海芳, 李聪, 姬永刚. 序约束下单向分类方差分析模型的Bayes变量选择[J]. 吉林大学学报(理学版), 2021, 59(5): 1093-1100.

SHI Haifang, LI Cong, JI Yonggang. Bayesian Variable Selection in One-Way Classified ANOVA Models under Order Constraints[J]. Journal of Jilin University Science Edition, 2021, 59(5): 1093-1100.

[1]	杜宇静, 姜丽萍. 序贯k-out-of-n系统在序约束下参数的假设检验[J]. 吉林大学学报(理学版), 2016, 54(03): 487-492.
[2]	史海芳, 姬永刚. 简单半序约束下多个正态总体分布参数的Bayes估计与等值检验[J]. J4, 2013, 51(01): 1-8.
[3]	齐芯, 王德辉. 序约束下AR(2)模型参数的拟似然估计[J]. J4, 2011, 49(03): 447-451.
[4]	赵世舜, 宋洋, 宋立新. 对称熵损失下两个指数总体均值的序约束估计[J]. J4, 2007, 45(01): 44-48.
[5]	李树有,, 史宁中, 赵世舜. 正态总体均值与标准差比在树序约束下的最大似然估计[J]. J4, 2006, 44(06): 893-897.
[6]	王晓光, 宋立新. 序约束下ARCH模型的最小二乘估计[J]. J4, 2005, 43(03): 287-294.