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

FindBook

Google Book

Amazon

博客來

Bayesian Hierarchical Difference-in-Differences Models.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Bayesian Hierarchical Difference-in-Differences Models./
作者:	Normington, James P.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
面頁冊數:	110 p.
附註:	Source: Dissertations Abstracts International, Volume: 81-09, Section: B.
Contained By:	Dissertations Abstracts International81-09B.
標題:	Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27672190
ISBN:	9781392588413

Bayesian Hierarchical Difference-in-Differences Models.
Normington, James P.

Bayesian Hierarchical Difference-in-Differences Models. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 110 p.

Source: Dissertations Abstracts International, Volume: 81-09, Section: B.

Thesis (Ph.D.)--University of Minnesota, 2019.

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

A popular method for estimating a causal treatment effect with observational data is the difference-in-differences (DiD) model. In this work, we extend the classical DiD model to the hierarchical context in which data cannot be matched at the most granular level. Our motivating example is an application to assess the impact of primary care redesign policy on diabetes outcomes in Minnesota, in which patient-level outcomes are not matched longitudinally, and thus the mean change in outcome, treatment(s), and covariates are measured at the clinic level. We propose a Bayesian hierarchical difference-in-differences (HDiD) model which estimates the treatment effect by regressing the treatment on a latent variable representing the mean change in group-level outcome. We first apply the HDiD model to measure the impact of primary care redesign on clinics in Minnesota from 2008 to 2012. We go on to present theoretical and empirical results showing that an HDiD model that fails to adjust for a particular class of confounding variables biases the treatment effect estimate. Motivated by the need for covariate adjustment, we propose and implement various approaches to perform variable selection using a structured Bayesian spike-and-slab model in the HDiD context, evaluating their properties with theoretical results and through simulation. We then conduct a cross-sectional analysis to measure the specific primary care services and resources which most strongly relate with improved diabetes outcomes in 2017. We conclude by introducing an additional variable selection approach which leverages the temporality of the HDiD context, as well as any structure in the predictors.

ISBN: 9781392588413Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Difference-in-differences models

Bayesian Hierarchical Difference-in-Differences Models.
LDR:02682nmm a2200301 4500 001 2348525
005 20220912135600.5
008 241004s2019 ||||||||||||||||| ||eng d
020 $a 9781392588413
035 $a (MiAaPQ)AAI27672190
035 $a AAI27672190
040 $a MiAaPQ $c MiAaPQ
100 1 $a Normington, James P. $3 3687891
245 1 0 $a Bayesian Hierarchical Difference-in-Differences Models.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 110 p.
500 $a Source: Dissertations Abstracts International, Volume: 81-09, Section: B.
500 $a Advisor: Lock, Eric F;Murray, Thomas A.
502 $a Thesis (Ph.D.)--University of Minnesota, 2019.
506 $a This item must not be sold to any third party vendors.
520 $a A popular method for estimating a causal treatment effect with observational data is the difference-in-differences (DiD) model. In this work, we extend the classical DiD model to the hierarchical context in which data cannot be matched at the most granular level. Our motivating example is an application to assess the impact of primary care redesign policy on diabetes outcomes in Minnesota, in which patient-level outcomes are not matched longitudinally, and thus the mean change in outcome, treatment(s), and covariates are measured at the clinic level. We propose a Bayesian hierarchical difference-in-differences (HDiD) model which estimates the treatment effect by regressing the treatment on a latent variable representing the mean change in group-level outcome. We first apply the HDiD model to measure the impact of primary care redesign on clinics in Minnesota from 2008 to 2012. We go on to present theoretical and empirical results showing that an HDiD model that fails to adjust for a particular class of confounding variables biases the treatment effect estimate. Motivated by the need for covariate adjustment, we propose and implement various approaches to perform variable selection using a structured Bayesian spike-and-slab model in the HDiD context, evaluating their properties with theoretical results and through simulation. We then conduct a cross-sectional analysis to measure the specific primary care services and resources which most strongly relate with improved diabetes outcomes in 2017. We conclude by introducing an additional variable selection approach which leverages the temporality of the HDiD context, as well as any structure in the predictors.
590 $a School code: 0130.
650 4 $a Biostatistics. $3 1002712
653 $a Difference-in-differences models
690 $a 0308
710 2 $a University of Minnesota. $b Biostatistics. $3 3179153
773 0 $t Dissertations Abstracts International $g 81-09B.
790 $a 0130
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=27672190