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

Latent variable models for longitudi...

Roy, Jason Allen.

Linked to FindBook

Google Book

Amazon

博客來

Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates./
Author:	Roy, Jason Allen.
Description:	106 p.
Notes:	Chair: Xihong Lin.
Contained By:	Dissertation Abstracts International61-07B.
Subject:	Biology, Biostatistics. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9977251
ISBN:	0599833106

Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates.
Roy, Jason Allen.

Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates. - 106 p.

Chair: Xihong Lin.

Thesis (Ph.D.)--University of Michigan, 2000.

In many studies the outcome of main interest cannot be measured by a single response. There is a great deal of literature dealing with such data for cross-sectional studies. However, this problem has not been well studied for longitudinal data. In this dissertation we propose latent variable models to handle this type of multivariate longitudinal data. At the first stage of the model we assume the observed outcomes measure the latent variable with error. The latent variable is then assumed associated with covariates through a linear mixed model. We extend this model to the situation where the probability of dropout is latent variable dependent, and hence non-ignorable. We first show how one can find maximum likelihood estimates when the covariates are completely observed. We then relax this assumption by allowing covariates to be missing due to unit dropout, which is often the case when there are time-varying covariates. Finally, we look at the missing covariate issue in more detail for the single outcome case. We carry out a bias analysis, comparing our proposed method with naive methods for handling the missing covariates. The Gibbs sampler for this model is developed to obtain Bayesian inference. Data from a national panel study on changes in methadone treatment practices are used throughout to illustrate the methodology.

ISBN: 0599833106Subjects--Topical Terms:

1018416
Biology, Biostatistics.

Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates.
LDR:02249nam 2200277 a 45 001 929294
005 20110427
008 110427s2000 eng d
020 $a 0599833106
035 $a (UnM)AAI9977251
035 $a AAI9977251
040 $a UnM $c UnM
100 1 $a Roy, Jason Allen. $3 1252778
245 1 0 $a Latent variable models for longitudinal data with multiple outcomes, informative dropouts and missing covariates.
300 $a 106 p.
500 $a Chair: Xihong Lin.
500 $a Source: Dissertation Abstracts International, Volume: 61-07, Section: B, page: 3357.
502 $a Thesis (Ph.D.)--University of Michigan, 2000.
520 $a In many studies the outcome of main interest cannot be measured by a single response. There is a great deal of literature dealing with such data for cross-sectional studies. However, this problem has not been well studied for longitudinal data. In this dissertation we propose latent variable models to handle this type of multivariate longitudinal data. At the first stage of the model we assume the observed outcomes measure the latent variable with error. The latent variable is then assumed associated with covariates through a linear mixed model. We extend this model to the situation where the probability of dropout is latent variable dependent, and hence non-ignorable. We first show how one can find maximum likelihood estimates when the covariates are completely observed. We then relax this assumption by allowing covariates to be missing due to unit dropout, which is often the case when there are time-varying covariates. Finally, we look at the missing covariate issue in more detail for the single outcome case. We carry out a bias analysis, comparing our proposed method with naive methods for handling the missing covariates. The Gibbs sampler for this model is developed to obtain Bayesian inference. Data from a national panel study on changes in methadone treatment practices are used throughout to illustrate the methodology.
590 $a School code: 0127.
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 Michigan. $3 777416
773 0 $t Dissertation Abstracts International $g 61-07B.
790 $a 0127
790 1 0 $a Lin, Xihong, $e advisor
791 $a Ph.D.
792 $a 2000
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=9977251