Sparse confidence sets for normal mean models

Sparse confidence sets for normal mean models
复制标题

正态平均模型的稀疏置信集

DOI:
10.1093/imaiai/iaad003
复制
发表时间:
2023
期刊:
Information and Inference: A Journal of the IMA
影响因子:
--
通讯作者:
Cheng, Guang
Cheng, Guang
中科院分区:
--
文献类型:
--
作者:
Ning, Yang;Cheng, Guang

文献摘要

参考文献

被引文献

相似文献

本文提出了一种在正态均值模型下构造一维未知稀疏参数置信集的新框架。所提出的置信度集的一个关键特征是它能够解释稀疏性,因此被称为辅助置信度集。这与经典的方法形成了鲜明的对比,例如Bonferroni置信度区间和其他基于重采样的过程,在这些方法中,稀疏性往往被忽略。具体地说,我们要求期望的稀疏置信度集满足以下两个条件:(I)在参数空间上一致地,覆盖概率高于预先指定的水平;(Ii)存在这样的随机子集,它保证预先指定的真负率来检测非零值。为了利用的稀疏性,我们允许对任何一个参数空间的置信度区间退化到一个点0。在这个新的框架下,我们首先考虑是否存在满足上述两个条件的稀疏置信集。为了解决这个问题,我们在一类适当的稀疏置信集上建立了非覆盖概率的非渐近极小极大下界。下界解释了稀疏性和最小信噪比在构造稀疏置信集中的作用。此外,在适当的信噪比条件下,提出了一种两阶段构造稀疏置信度集的方法。为了评估最优性,所提出的稀疏置信集被证明达到某一适当定义的风险函数的极小极大下界,直到一个常数因子。最后,我们开发了一种对未知稀疏性的自适应过程。通过数值研究对理论结果进行了验证。
In this paper, we propose a new framework to construct confidence sets for a-dimensional unknown sparse parameterunder the normal mean model. A key feature of the proposed confidence set is its capability to account for the sparsity of, thus named assparseconfidence set. This is in sharp contrast with the classical methods, such as the Bonferroni confidence intervals and other resampling-based procedures, where the sparsity ofis often ignored. Specifically, we require the desired sparse confidence set to satisfy the following two conditions: (i) uniformly over the parameter space, the coverage probability foris above a pre-specified level; (ii) there exists a random subsetofsuch thatguarantees the pre-specified true negative rate for detecting non-zero’s. To exploit the sparsity of, we allow the confidence interval forto degenerate to a single point 0 for any. Under this new framework, we first consider whether there exist sparse confidence sets that satisfy the above two conditions. To address this question, we establish a non-asymptotic minimax lower bound for the non-coverage probability over a suitable class of sparse confidence sets. The lower bound deciphers the role of sparsity and minimum signal-to-noise ratio (SNR) in the construction of sparse confidence sets. Furthermore, under suitable conditions on the SNR, a two-stage procedure is proposed to construct a sparse confidence set. To evaluate the optimality, the proposed sparse confidence set is shown to attain a minimax lower bound of some properly defined risk function up to a constant factor. Finally, we develop an adaptive procedure to the unknown sparsity. Numerical studies are conducted to verify the theoretical results.
选择具有更强确定符号能力的调整置信区间
DOI: --
发表时间: 2013
期刊:
影响因子: --
作者:
A. Weinstein;William Fithian;Y. Benjamini
通讯作者: Y. Benjamini
DOI: 10.1111/biom.13222
发表时间: 2020
期刊: Biometrics
影响因子: 1.9
作者:
Haibing Zhao;Xinping Cui
通讯作者: Xinping Cui
DOI: 10.1214/16-aos1483
发表时间: 2017-12-01
影响因子: 4.5
作者:
Ning, Yang;Zhao, Tianqi;Liu, Han
通讯作者: Liu, Han
DOI: 10.1214/10-ejs589
发表时间: 2010-01-01
影响因子: 1.1
作者:
Ingster, Yuri I.;Tsybakov, Alexandre B.;Verzelen, Nicolas
通讯作者: Verzelen, Nicolas
DOI: 10.1214/18-sts661
发表时间: 2018-08-01
影响因子: 5.7
作者:
Neykov, Matey;Ning, Yang;Liu, Han
通讯作者: Liu, Han