Bayesian Fusion Estimation via t Shrinkage

Bayesian Fusion Estimation via t Shrinkage
复制标题

通过 t 收缩进行贝叶斯融合估计

DOI:
10.1007/s13171-019-00177-0
复制
发表时间:
2020
期刊:
Sankhya A
影响因子:
--
通讯作者:
Cheng, Guang
Cheng, Guang
中科院分区:
--
文献类型:
--
作者:
Song, Qifan;Cheng, Guang

文献摘要

参考文献

被引文献

相似文献

收缩先验在许多数据分析中取得了巨大的成功,然而,它的应用主要集中在稀疏参数的贝叶斯建模上。在这项工作中,我们将应用贝叶斯收缩来模拟具有未知块结构的高维参数。我们建议对连续参数条目的差异施加重尾收缩先验,例如tprior,这样的融合先验将连续差异缩小到零,从而导致后验阻塞。与实现拉普拉斯融合先验的传统贝叶斯融合LASSO相比,融合先验具有更强的收缩效应,并具有良好的后验一致性。仿真研究和实际数据分析表明,该融合方法比频率估计方法和贝叶斯拉普拉斯融合先验方法具有更好的性能。该融合策略被进一步发展以进行贝叶斯聚类分析,我们的模拟表明,所提出的算法优于经典的狄利克雷过程建模。
Shrinkage prior has gained great successes in many data analysis, however, its applications mostly focus on the Bayesian modeling of sparse parameters. In this work, we will apply Bayesian shrinkage to model high dimensional parameter that possesses an unknown blocking structure. We propose to impose heavy-tail shrinkage prior, e.g.,tprior, on the differences of successive parameter entries, and such a fusion prior will shrink successive differences towards zero and hence induce posterior blocking. Comparing to conventional Bayesian fused LASSO which implements Laplace fusion prior,tfusion prior induces stronger shrinkage effect and enjoys a nice posterior consistency property. Simulation studies and real data analyses show thattfusion has superior performance to the frequentist fusion estimator and Bayesian Laplace fusion prior. Thistfusion strategy is further developed to conduct a Bayesian clustering analysis, and our simulations show that the proposed algorithm compares favorably to classical Dirichlet process modeling.
DOI: 10.1080/01621459.2018.1529595
发表时间: 2019-04-10
影响因子: 3.7
作者:
Li, Furong;Sang, Huiyan
通讯作者: Sang, Huiyan
A. R. Barron 受邀讨论:贝叶斯性能的信息理论表征以及参数和非参数问题中先验的选择
DOI: --
发表时间: 1998
期刊:
影响因子: --
作者:
T. Sweeting
通讯作者: T. Sweeting
动态二元网络的贝叶斯融合套索回归
DOI: 10.1080/10618600.2017.1341323
发表时间: 2017
影响因子: 2.4
作者:
Betancourt, Brenda;Rodríguez, Abel;Boyd, Naomi
通讯作者: Boyd, Naomi
无限维贝叶斯统计中的错误指定
DOI: 10.1214/009053606000000029
发表时间: 2006
影响因子: 4.5
作者:
B. Kleijn;W. A.;Van der Vaart
通讯作者: Van der Vaart
DOI: 10.1198/016214506000000735
发表时间: 2006-12-01
影响因子: 3.7
作者:
Zou, Hui
通讯作者: Zou, Hui