國立虎尾科技大學 |

Bayesian Semiparametric Measurement Error Models : = Estimation, Variable Selection and Fast Algorithms.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Bayesian Semiparametric Measurement Error Models :/
其他題名:	Estimation, Variable Selection and Fast Algorithms.
作者:	Liu, Chang.
面頁冊數:	1 online resource (251 pages)
附註:	Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
Contained By:	Dissertation Abstracts International78-10B(E).
標題:	Statistics. -
電子資源:	click for full text (PQDT)
ISBN:	9781369823011

Bayesian Semiparametric Measurement Error Models : = Estimation, Variable Selection and Fast Algorithms.
Liu, Chang.

Bayesian Semiparametric Measurement Error Models :Estimation, Variable Selection and Fast Algorithms. - 1 online resource (251 pages)

Source: Dissertation Abstracts International, Volume: 78-10(E), Section: B.

Thesis (Ph.D.)

Includes bibliographical references

This thesis presents Bayesian estimation, variable selection and fast algorithms for semiparametric measurement error models. We will extend the existing Bayesian estimation procedures for measurement error models to allow for heteroscedasctic regression errors from a normal distribution and further to relax the normality assumptio. Bayesian penalized splines will be used to estimate the variance functions and Dirichlet process mixtures models will be applied for non-normal regression errors. Variable selection procedures for this class of model will also be considered. We will apply Bayesian adaptive LASSO and group LASSO to jointly select and estimate the parametric and nonparametric components in a semiparametric model in the presence of measurement error, either in the linear or nonlinear covariate. Some extensions will be discussed to account for heteroscedasticity. Several computing strategies will be proposed to deal with high-dimensional data. To address the computational issues associated with Bayesian posterior inference, we will propose several mean field variational Bayes approximation algorithms for estimation and variable selection algorithms. Simulation studies will be conducted to evaluate each algorithm. The proposed algorithms will be applied to data on phthalate exposure, diet and semen quality.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9781369823011Subjects--Topical Terms:

556824
Statistics.
Index Terms--Genre/Form:

554714
Electronic books.

Bayesian Semiparametric Measurement Error Models : = Estimation, Variable Selection and Fast Algorithms.
LDR:02631ntm a2200337Ki 4500 001 911545
005 20180529094435.5
006 m o u
007 cr mn||||a|a||
008 190606s2017 xx obm 000 0 eng d
020 $a 9781369823011
035 $a (MiAaPQ)AAI10268909
035 $a (MiAaPQ)rochester:11409
035 $a AAI10268909
040 $a MiAaPQ $b eng $c MiAaPQ
099 $a TUL $f hyy $c available through World Wide Web
100 1 $a Liu, Chang. $3 1179760
245 1 0 $a Bayesian Semiparametric Measurement Error Models : $b Estimation, Variable Selection and Fast Algorithms.
264 0 $c 2017
300 $a 1 online resource (251 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: Dissertation Abstracts International, Volume: 78-10(E), Section: B.
500 $a Adviser: Sally W. Thurston.
502 $a Thesis (Ph.D.) $c University of Rochester $d 2017.
504 $a Includes bibliographical references
520 $a This thesis presents Bayesian estimation, variable selection and fast algorithms for semiparametric measurement error models. We will extend the existing Bayesian estimation procedures for measurement error models to allow for heteroscedasctic regression errors from a normal distribution and further to relax the normality assumptio. Bayesian penalized splines will be used to estimate the variance functions and Dirichlet process mixtures models will be applied for non-normal regression errors. Variable selection procedures for this class of model will also be considered. We will apply Bayesian adaptive LASSO and group LASSO to jointly select and estimate the parametric and nonparametric components in a semiparametric model in the presence of measurement error, either in the linear or nonlinear covariate. Some extensions will be discussed to account for heteroscedasticity. Several computing strategies will be proposed to deal with high-dimensional data. To address the computational issues associated with Bayesian posterior inference, we will propose several mean field variational Bayes approximation algorithms for estimation and variable selection algorithms. Simulation studies will be conducted to evaluate each algorithm. The proposed algorithms will be applied to data on phthalate exposure, diet and semen quality.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Statistics. $3 556824
655 7 $a Electronic books. $2 local $3 554714
690 $a 0463
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a University of Rochester. $b Medicine and Dentistry. $3 1183441
773 0 $t Dissertation Abstracts International $g 78-10B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10268909 $z click for full text (PQDT)