東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Regularization for Variable Selectio...

Yang, Yang.

Linked to FindBook

Google Book

Amazon

博客來

Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application./
Author:	Yang, Yang.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
Description:	108 p.
Notes:	Source: Dissertations Abstracts International, Volume: 80-04, Section: B.
Contained By:	Dissertations Abstracts International80-04B.
Subject:	Biostatistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10840985
ISBN:	9780438455221

Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application.
Yang, Yang.

Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 108 p.

Source: Dissertations Abstracts International, Volume: 80-04, Section: B.

Thesis (Ph.D.)--State University of New York at Buffalo, 2018.

This item must not be sold to any third party vendors.

Regularized likelihood has been proven to be effective in variable selection. This method has been well developed theoretically and computationally in the past two decades. However, two major problems still exist in practice. One is caused by non-ignorable missing data. The sensitivity of missing mechanism assumption sets obstacles to select variables of interest. The other one is that the normality of response variable is rare in practice. Clinical data are usually distributed asymmetrically and sometimes finitely, preventing direct application of penalized likelihood. Research in this dissertation is driven by the need for statistical tools to apply regularized likelihood in practice through solving the two problems stated above. This dissertation is organized as follow. An introduction of our research topic is given in Chapter 1. We focus on a flexible and generally applicable missing data mechanism, which contains both ignorable and nonignorable missing data mechanism assumptions. In Chapter 2, we show that how to achieve variable selection purpose through incorpating this missing data mechanism and a pseudo likelihood funciton together when missing data, especially nonignorable missing data, exists. The computational algorithm used to optimize the objective function is also developed. Theoretical properties for variable selection consistency is discussed in Chapter 3 and all technical proofs are included in Chapter 8. Tuning paramter selection is another important topic for variable selection as it controls the balance between complexity of the model and prediction accuracy. In Chapter 4, we not only expolore and extend currently existed tuning parameter selection method, but also propose a new stability-based tuning parameter selection method to select the optimal tuning parameter. In Chapter 5 and 6, we explore our method through comprehensive simulation and real data studies. We conclude this dissertation with a discussion in Chapter 7.

ISBN: 9780438455221Subjects--Topical Terms:

1002712
Biostatistics.

Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application.
LDR:03121nmm a2200337 4500 001 2208460
005 20191021073442.5
008 201008s2018 ||||||||||||||||| ||eng d
020 $a 9780438455221
035 $a (MiAaPQ)AAI10840985
035 $a (MiAaPQ)buffalo:15914
035 $a AAI10840985
040 $a MiAaPQ $c MiAaPQ
100 1 $a Yang, Yang. $3 1030201
245 1 0 $a Regularization for Variable Selection with Flexible Missingness Mechanism: Methodology, Algorithm and Application.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 108 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-04, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Zhao, Jiwei.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2018.
506 $a This item must not be sold to any third party vendors.
520 $a Regularized likelihood has been proven to be effective in variable selection. This method has been well developed theoretically and computationally in the past two decades. However, two major problems still exist in practice. One is caused by non-ignorable missing data. The sensitivity of missing mechanism assumption sets obstacles to select variables of interest. The other one is that the normality of response variable is rare in practice. Clinical data are usually distributed asymmetrically and sometimes finitely, preventing direct application of penalized likelihood. Research in this dissertation is driven by the need for statistical tools to apply regularized likelihood in practice through solving the two problems stated above. This dissertation is organized as follow. An introduction of our research topic is given in Chapter 1. We focus on a flexible and generally applicable missing data mechanism, which contains both ignorable and nonignorable missing data mechanism assumptions. In Chapter 2, we show that how to achieve variable selection purpose through incorpating this missing data mechanism and a pseudo likelihood funciton together when missing data, especially nonignorable missing data, exists. The computational algorithm used to optimize the objective function is also developed. Theoretical properties for variable selection consistency is discussed in Chapter 3 and all technical proofs are included in Chapter 8. Tuning paramter selection is another important topic for variable selection as it controls the balance between complexity of the model and prediction accuracy. In Chapter 4, we not only expolore and extend currently existed tuning parameter selection method, but also propose a new stability-based tuning parameter selection method to select the optimal tuning parameter. In Chapter 5 and 6, we explore our method through comprehensive simulation and real data studies. We conclude this dissertation with a discussion in Chapter 7.
590 $a School code: 0656.
650 4 $a Biostatistics. $3 1002712
650 4 $a Mathematics. $3 515831
650 4 $a Statistics. $3 517247
690 $a 0308
690 $a 0405
690 $a 0463
710 2 $a State University of New York at Buffalo. $b Biostatistics. $3 2099967
773 0 $t Dissertations Abstracts International $g 80-04B.
790 $a 0656
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10840985