
arXiv (Cornell University), Год журнала: 2021, Номер unknown
Опубликована: Янв. 1, 2021
Large-scale modern data often involves estimation and testing for high-dimensional unknown parameters. It is desirable to identify the sparse signals, ``the needles in haystack'', with accuracy false discovery control. However, unprecedented complexity heterogeneity structure require new machine learning tools effectively exploit commonalities robustly adjust both sparsity heterogeneity. In addition, estimates parameters lack uncertainty quantification. this paper, we propose a novel Spike-and-Nonparametric mixture prior (SNP) -- spike promote nonparametric capture signals. contrast state-of-the-art methods, proposed methods solve problem at once several merits: 1) an accurate estimation; 2) point shrinkage/soft-thresholding property; 3) credible intervals quantification; 4) optimal multiple procedure that controls rate. Our method exhibits promising empirical performance on simulated gene expression case study.
Язык: Английский