國立虎尾科技大學 |

Bayesian Methods for Variable Selection.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Bayesian Methods for Variable Selection./
作者:	Porwal, Anupreet.
面頁冊數:	1 online resource (194 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
Contained By:	Dissertations Abstracts International85-03B.
標題:	Applied mathematics. -
電子資源:	click for full text (PQDT)
ISBN:	9798380328388

Bayesian Methods for Variable Selection.
Porwal, Anupreet.

Bayesian Methods for Variable Selection. - 1 online resource (194 pages)

Source: Dissertations Abstracts International, Volume: 85-03, Section: B.

Thesis (Ph.D.)--University of Washington, 2023.

Includes bibliographical references

Choosing a statistical model and accounting for uncertainty about this choice are important parts of the scientific process and are required for common statistical tasks such as parameter estimation, interval estimation, statistical inference, point prediction and interval prediction. A canonical example is the variable selection problem in a linear regression model. Many ways of doing this have been proposed, including Bayesian and penalized regression methods. Each of these proposals has advantages and disadvantages, and the trade-offs are not always well understood.In this dissertation, we first compare 21 popular existing methods via an extensive simulation study based on a wide range of real datasets. We found that three adaptive Bayesian model averaging (BMA) methods performed best across all the statistical tasks. Subsequently, we also investigate the effect of model space priors on model inference under the BMA framework. For this study, we consider eight reference model space priors used in the literature and three adaptive parameter priors recommended by the previous study.Additionally, we proposed a novel objective prior based on Power-expected-posterior priors for generalized linear models that relies on a Laplace expansion of the likelihood of the imaginary training sample. We investigate both asymptotic and finite-sample properties of the procedures, showing that they are both asymptotically and intrinsically consistent, and that their performance is superior to that of alternative approaches in the literature especially for heavy tailed versions of the priors.Finally, we propose a framework that unifies the two Bayesian paradigms of inducing sparsity namely (mixture of) g-priors and continuous shrinkage priors. The mixture of g-priors use a single shrinkage parameter across all predictors included in the model, incorporate correlation structure of covariates into the priors and allows for model selection, however suffers from Conditional Lindley Paradox (CLP). Continuous shrinkage priors like the Horseshoe prior, on the other hand, allow for a different shrinkage parameter for each coefficient however does not perform model selection. We propose global local-g priors that borrows strength of the two paradigms and allows differential shrinkage across predictors while performing model selection. Additionally, we propose Dirichlet Process (DP) block-g priors that allows combining g-priors and Bayesian non-parametric tools to incorporate correlation structure in the priors as well as adaptively identify and cluster predictors with varying degrees of relevance using a different shrinkage parameter for different clusters. We show empirically and theoretically that our proposed priors avoid CLP while still performing competitively or superior to existing methods in terms of model selection, parameter estimation and prediction.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798380328388Subjects--Topical Terms:

1069907
Applied mathematics.
Subjects--Index Terms:

Bayesian model averagingIndex Terms--Genre/Form:

554714
Electronic books.

Bayesian Methods for Variable Selection.
LDR:04295ntm a22004097 4500 001 1147541
005 20240909103808.5
006 m o d
007 cr bn ---uuuuu
008 250605s2023 xx obm 000 0 eng d
020 $a 9798380328388
035 $a (MiAaPQ)AAI30635571
035 $a AAI30635571
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Porwal, Anupreet. $3 1473298
245 1 0 $a Bayesian Methods for Variable Selection.
264 0 $c 2023
300 $a 1 online resource (194 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 85-03, Section: B.
500 $a Advisor: Rodriguez, Abel;Raftery, Adrian E.
502 $a Thesis (Ph.D.)--University of Washington, 2023.
504 $a Includes bibliographical references
520 $a Choosing a statistical model and accounting for uncertainty about this choice are important parts of the scientific process and are required for common statistical tasks such as parameter estimation, interval estimation, statistical inference, point prediction and interval prediction. A canonical example is the variable selection problem in a linear regression model. Many ways of doing this have been proposed, including Bayesian and penalized regression methods. Each of these proposals has advantages and disadvantages, and the trade-offs are not always well understood.In this dissertation, we first compare 21 popular existing methods via an extensive simulation study based on a wide range of real datasets. We found that three adaptive Bayesian model averaging (BMA) methods performed best across all the statistical tasks. Subsequently, we also investigate the effect of model space priors on model inference under the BMA framework. For this study, we consider eight reference model space priors used in the literature and three adaptive parameter priors recommended by the previous study.Additionally, we proposed a novel objective prior based on Power-expected-posterior priors for generalized linear models that relies on a Laplace expansion of the likelihood of the imaginary training sample. We investigate both asymptotic and finite-sample properties of the procedures, showing that they are both asymptotically and intrinsically consistent, and that their performance is superior to that of alternative approaches in the literature especially for heavy tailed versions of the priors.Finally, we propose a framework that unifies the two Bayesian paradigms of inducing sparsity namely (mixture of) g-priors and continuous shrinkage priors. The mixture of g-priors use a single shrinkage parameter across all predictors included in the model, incorporate correlation structure of covariates into the priors and allows for model selection, however suffers from Conditional Lindley Paradox (CLP). Continuous shrinkage priors like the Horseshoe prior, on the other hand, allow for a different shrinkage parameter for each coefficient however does not perform model selection. We propose global local-g priors that borrows strength of the two paradigms and allows differential shrinkage across predictors while performing model selection. Additionally, we propose Dirichlet Process (DP) block-g priors that allows combining g-priors and Bayesian non-parametric tools to incorporate correlation structure in the priors as well as adaptively identify and cluster predictors with varying degrees of relevance using a different shrinkage parameter for different clusters. We show empirically and theoretically that our proposed priors avoid CLP while still performing competitively or superior to existing methods in terms of model selection, parameter estimation and prediction.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Applied mathematics. $3 1069907
650 4 $a Mathematics. $3 527692
650 4 $a Statistics. $3 556824
653 $a Bayesian model averaging
653 $a Default priors
653 $a Generalised linear models
653 $a Linear regression
653 $a Model selection
653 $a Zellner's g priors
655 7 $a Electronic books. $2 local $3 554714
690 $a 0463
690 $a 0405
690 $a 0364
710 2 $a University of Washington. $b Statistics. $3 1182938
710 2 $a ProQuest Information and Learning Co. $3 1178819
773 0 $t Dissertations Abstracts International $g 85-03B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30635571 $z click for full text (PQDT)