東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Problems in Variable Selection: Fals...

Lin, Buyu.

FindBook

Google Book

Amazon

博客來

Problems in Variable Selection: False Discovery Rate Control and Variational Inference.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Problems in Variable Selection: False Discovery Rate Control and Variational Inference./
作者:	Lin, Buyu.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2024,
面頁冊數:	316 p.
附註:	Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
Contained By:	Dissertations Abstracts International85-12B.
標題:	Statistics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31294278
ISBN:	9798382777672

Problems in Variable Selection: False Discovery Rate Control and Variational Inference.
Lin, Buyu.

Problems in Variable Selection: False Discovery Rate Control and Variational Inference. - Ann Arbor : ProQuest Dissertations & Theses, 2024 - 316 p.

Source: Dissertations Abstracts International, Volume: 85-12, Section: B.

Thesis (Ph.D.)--Harvard University, 2024.

Variable selection plays a key role in modern high-dimensional statistics. This dissertation provides a comprehensive survey of theory and methods developed by the author and collaborators, with a specific focus on two areas: false discovery rate (FDR) control and Bayesian variable selection. In the domain of FDR control, we introduce a data-splitting method to asymptotically control the FDR while maintaining a high power. Furthermore, a Multiple Data Splitting (MDS) method is proposed to stabilize the selection result and boost the power. In Chapter 1, we apply both DS and MDS to the generalized linear models, which appear to be more robust in finite-sample cases compared to existing methods. Chapter 2 provides some following discussions regarding the proposed method and Chapter 3 compares the power of the proposed method with two existing methods: the model-X knockoff and Gaussian mirror. In terms of the Bayesian variable selection, the posterior is typically high-dimensional and analytically intractable. Exact inference methods based on sampling, such as Markov Chain Monte Carlo (MCMC), can encounter challenges related to mixing. Variational inference has emerged as an attractive alternative for approximating the posterior distribution. By recasting the sampling problem as an optimization problem, variational inference can significantly reduce computational time. In chapter 4, we apply the variational inference to group variable selection with spike-and-slab prior and propose an efficient parameter-expanded coordinate-ascent algorithm to obtain the optimal variational Bayes approximation. The proposed method has demonstrated good performance in both simulations and a real data example.

ISBN: 9798382777672Subjects--Topical Terms:

517247
Statistics.
Subjects--Index Terms:

Multiple Data Splitting

Problems in Variable Selection: False Discovery Rate Control and Variational Inference.
LDR:02887nmm a2200385 4500 001 2399339
005 20240909103820.5
006 m o d
007 cr#unu||||||||
008 251215s2024 ||||||||||||||||| ||eng d
020 $a 9798382777672
035 $a (MiAaPQ)AAI31294278
035 $a AAI31294278
040 $a MiAaPQ $c MiAaPQ
100 1 $a Lin, Buyu. $0 (orcid)0009-0000-2569-3753 $3 3769308
245 1 0 $a Problems in Variable Selection: False Discovery Rate Control and Variational Inference.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 316 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-12, Section: B.
500 $a Advisor: Liu, Jun.
502 $a Thesis (Ph.D.)--Harvard University, 2024.
520 $a Variable selection plays a key role in modern high-dimensional statistics. This dissertation provides a comprehensive survey of theory and methods developed by the author and collaborators, with a specific focus on two areas: false discovery rate (FDR) control and Bayesian variable selection. In the domain of FDR control, we introduce a data-splitting method to asymptotically control the FDR while maintaining a high power. Furthermore, a Multiple Data Splitting (MDS) method is proposed to stabilize the selection result and boost the power. In Chapter 1, we apply both DS and MDS to the generalized linear models, which appear to be more robust in finite-sample cases compared to existing methods. Chapter 2 provides some following discussions regarding the proposed method and Chapter 3 compares the power of the proposed method with two existing methods: the model-X knockoff and Gaussian mirror. In terms of the Bayesian variable selection, the posterior is typically high-dimensional and analytically intractable. Exact inference methods based on sampling, such as Markov Chain Monte Carlo (MCMC), can encounter challenges related to mixing. Variational inference has emerged as an attractive alternative for approximating the posterior distribution. By recasting the sampling problem as an optimization problem, variational inference can significantly reduce computational time. In chapter 4, we apply the variational inference to group variable selection with spike-and-slab prior and propose an efficient parameter-expanded coordinate-ascent algorithm to obtain the optimal variational Bayes approximation. The proposed method has demonstrated good performance in both simulations and a real data example.
590 $a School code: 0084.
650 4 $a Statistics. $3 517247
650 4 $a Endocrinology. $3 610914
650 4 $a Biostatistics. $3 1002712
653 $a Multiple Data Splitting
653 $a False discovery rate
653 $a Markov Chain Monte Carlo
653 $a Variational inference
653 $a FDR control
690 $a 0463
690 $a 0409
690 $a 0308
710 2 $a Harvard University. $b Statistics. $3 2093214
773 0 $t Dissertations Abstracts International $g 85-12B.
790 $a 0084
791 $a Ph.D.
792 $a 2024
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31294278