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

Practical issues in Gibbs sampler im...

Cowles, Mary Kathryn.

FindBook

Google Book

Amazon

博客來

Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data./
作者:	Cowles, Mary Kathryn.
面頁冊數:	148 p.
附註:	Co-Major Advisers: Thomas A. Louis; Bradley P. Carlin.
Contained By:	Dissertation Abstracts International55-07B.
標題:	Biology, Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9433042

Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data.
Cowles, Mary Kathryn.

Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data. - 148 p.

Co-Major Advisers: Thomas A. Louis; Bradley P. Carlin.

Thesis (Ph.D.)--University of Minnesota, 1994.

In addition, software is developed to assist the applied user of sampling-based methods with convergence diagnosis and output analysis.Subjects--Topical Terms:

1018416
Biology, Biostatistics.

Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data.
LDR:02686nam 2200289 a 45 001 932665
005 20110505
008 110505s1994 eng d
035 $a (UnM)AAI9433042
035 $a AAI9433042
040 $a UnM $c UnM
100 1 $a Cowles, Mary Kathryn. $3 1256408
245 1 0 $a Practical issues in Gibbs sampler implementation with application to Bayesian hierarchical modeling of clinical trial data.
300 $a 148 p.
500 $a Co-Major Advisers: Thomas A. Louis; Bradley P. Carlin.
500 $a Source: Dissertation Abstracts International, Volume: 55-07, Section: B, page: 2481.
502 $a Thesis (Ph.D.)--University of Minnesota, 1994.
520 $a In addition, software is developed to assist the applied user of sampling-based methods with convergence diagnosis and output analysis.
520 $a In the Lung Health Study, compliance with the use of inhaled medication is assessed at each follow-up visit both by self-report and by weighing the used medication canisters. One or both of these assessments are missing if the participant fails to attend the visit or to return all canisters. We use Gibbs sampling with data augmentation to implement a Bayesian hierarchical model for Lung Health Study inhaler compliance that allows (a) assessment of the relationships among visit attendance, canister return, self-reported compliance level, and canister weight compliance, and (b) determination of demographic, physiological, and behavioral predictors of compliance. The model is a longitudinal extension of the Tobit models used in econometrics to deal with partially unobservable data, and incorporates (a) individual-level random effects to account for correlations among repeated measures on the same participant, (b) an informative prior on the missing data to account for a nonignorable missing data mechanism, (c) multivariate response variables, and (d) one ordinal response variable. An efficient multivariate Hastings update step is devised to accelerate the convergence of the Gibbs sampler for this model. Since the complexity of the models entertained thwarts any efforts to determine useful theoretical bounds for the required number of iterations of the Gibbs sampler, the performance of eight methods of diagnosing convergence is compared for simple models for the purpose of selecting a method for use in the final analysis of compliance data.
590 $a School code: 0130.
650 4 $a Biology, Biostatistics. $3 1018416
650 4 $a Statistics. $3 517247
690 $a 0308
690 $a 0463
710 2 0 $a University of Minnesota. $3 676231
773 0 $t Dissertation Abstracts International $g 55-07B.
790 $a 0130
790 1 0 $a Carlin, Bradley P., $e advisor
790 1 0 $a Louis, Thomas A., $e advisor
791 $a Ph.D.
792 $a 1994
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9433042