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

Planned Missing Data Designs for Cau...

Su, Dan.

FindBook

Google Book

Amazon

博客來

Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation./
作者:	Su, Dan.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2019,
面頁冊數:	80 p.
附註:	Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
Contained By:	Dissertations Abstracts International80-12B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13883768
ISBN:	9781392190241

Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation.
Su, Dan.

Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation. - Ann Arbor : ProQuest Dissertations & Theses, 2019 - 80 p.

Source: Dissertations Abstracts International, Volume: 80-12, Section: B.

Thesis (Ph.D.)--The University of Wisconsin - Madison, 2019.

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

Planned missing data designs in large surveys can efficiently reduce respondents' burden and lower the cost associated with data collection, without cutting down on the questionnaire items. If the missing data are not appropriately planned, it results bias in descriptive and potential causal parameter estimates. For a fixed sample size, the extend of bias depends on the three major characteristics of design and data: the missing percentage, the overlap percentage (i.e., the portion of the cases where two items are observed jointly), and the distributions of variables. My first two simulation studies investigate how the bias in marginal means, correlations and regression coefficients depends on the chosen planned missing data designs and the related characteristics.Even if a planned missing data design allows researchers to recover parameters of interest without bias, an incorrect choice of covariates at the imputation stage might actually introduce bias. For example, if the missing data pattern of a specific form or booklet causes context effects on an auxiliary variable that is used for imputing missing values, bias can be introduced. Thus, including all measured variables in the imputation model is not necessary a good strategy and, given the huge number of items in large surveys, frequently is problematic. The question then is, how should researchers select the imputation variables to obtain valid parameter estimates? The simulation studies investigate which variables not necessary or should not be included in the imputation model. Graphical models provide the theoretical basis for my simulations and explanations.

ISBN: 9781392190241Subjects--Topical Terms:

517247
Statistics.

Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation.
LDR:02715nmm a2200313 4500 001 2208563
005 20191021073504.5
008 201008s2019 ||||||||||||||||| ||eng d
020 $a 9781392190241
035 $a (MiAaPQ)AAI13883768
035 $a (MiAaPQ)wisc:16176
035 $a AAI13883768
040 $a MiAaPQ $c MiAaPQ
100 1 $a Su, Dan. $3 2095688
245 1 0 $a Planned Missing Data Designs for Causal Inference in Large Surveys: Design and Imputation.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2019
300 $a 80 p.
500 $a Source: Dissertations Abstracts International, Volume: 80-12, Section: B.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Kaplan, David.
502 $a Thesis (Ph.D.)--The University of Wisconsin - Madison, 2019.
506 $a This item must not be sold to any third party vendors.
520 $a Planned missing data designs in large surveys can efficiently reduce respondents' burden and lower the cost associated with data collection, without cutting down on the questionnaire items. If the missing data are not appropriately planned, it results bias in descriptive and potential causal parameter estimates. For a fixed sample size, the extend of bias depends on the three major characteristics of design and data: the missing percentage, the overlap percentage (i.e., the portion of the cases where two items are observed jointly), and the distributions of variables. My first two simulation studies investigate how the bias in marginal means, correlations and regression coefficients depends on the chosen planned missing data designs and the related characteristics.Even if a planned missing data design allows researchers to recover parameters of interest without bias, an incorrect choice of covariates at the imputation stage might actually introduce bias. For example, if the missing data pattern of a specific form or booklet causes context effects on an auxiliary variable that is used for imputing missing values, bias can be introduced. Thus, including all measured variables in the imputation model is not necessary a good strategy and, given the huge number of items in large surveys, frequently is problematic. The question then is, how should researchers select the imputation variables to obtain valid parameter estimates? The simulation studies investigate which variables not necessary or should not be included in the imputation model. Graphical models provide the theoretical basis for my simulations and explanations.
590 $a School code: 0262.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 $a The University of Wisconsin - Madison. $b Educational Psychology. $3 3170079
773 0 $t Dissertations Abstracts International $g 80-12B.
790 $a 0262
791 $a Ph.D.
792 $a 2019
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13883768