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 到查詢結果 [ null ]

切換: 標籤 | MARC模式 | ISBD

Contributions to Parametric and Semi...

Xu, Yifei,

FindBook

Google Book

Amazon

博客來

Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences /

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences // Yifei Xu.
作者:	Xu, Yifei,
面頁冊數:	1 electronic resource (121 pages)
附註:	Source: Dissertations Abstracts International, Volume: 85-10, Section: B.
Contained By:	Dissertations Abstracts International85-10B.
標題:	Biostatistics. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31143197
ISBN:	9798382221908

Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences /
Xu, Yifei,

Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences /Yifei Xu. - 1 electronic resource (121 pages)

Source: Dissertations Abstracts International, Volume: 85-10, Section: B.

This research work commences by discussing two commonly used fit indices: root mean squared error of approximation (RMSEA) and goodness-of-fit index (GFI), which are employed to assess the goodness of fit of model to the data. Chi-square distribution in the traditional RMSEA may vary with different estimation methods or types of data, leading to inaccurate RMSEA values, therefore, we propose an alternative RMSEA that is not sensitive to changes in degrees of freedom. Further-more, the traditional GFI tends to underperform on dichotomous data, particularly with small sample sizes. The method to mitigate the effect of a mis-specified fit function in the real world is integrated with a modified version of the GFI, namely, the corrected GFI (CGFI). A simulation study is employed to compare the performance of improved RMSEA and GFI with the previously modified RMSEA and GFI, respectively. Furthermore, as there is no previous research considering the non-linear relationship between confounding variables, manifest variables, and latent variables, the proposed SSEM introduces nonparametric link functions, assumed to be monotone/non-monotone, to classical SEM instead of constructing a mixture multivariate model based on latent variables. Four types of regressions are used to predict the functional forms in the simulation study. Two metabolic syndrome datasets are used to demonstrate the application of the methods.

English

ISBN: 9798382221908Subjects--Topical Terms:

1002712
Biostatistics.
Subjects--Index Terms:

Categorical data

Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences /
LDR:02975nmm a22004453i 4500 001 2428996
005 20250428055541.5
006 m o d
007 cr|nu||||||||
008 261011s2024 miu||||||m |||||||eng d
020 $a 9798382221908
035 $a (MiAaPQD)AAI31143197
035 $a AAI31143197
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Xu, Yifei, $e author. $0 (orcid)0009-0006-5057-7352 $3 3816570
245 1 0 $a Contributions to Parametric and Semiparametric Structural Equation Modeling Methods With Applications in the Biomedical Sciences / $c Yifei Xu.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2024
300 $a 1 electronic resource (121 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: Dissertations Abstracts International, Volume: 85-10, Section: B.
500 $a Advisors: Makambi, Kepher; Yuan, Ao Committee members: Korostyshevskiy, Valeriy; Dash, Chiranjeev.
502 $b Ph.D. $c Georgetown University $d 2024.
520 $a This research work commences by discussing two commonly used fit indices: root mean squared error of approximation (RMSEA) and goodness-of-fit index (GFI), which are employed to assess the goodness of fit of model to the data. Chi-square distribution in the traditional RMSEA may vary with different estimation methods or types of data, leading to inaccurate RMSEA values, therefore, we propose an alternative RMSEA that is not sensitive to changes in degrees of freedom. Further-more, the traditional GFI tends to underperform on dichotomous data, particularly with small sample sizes. The method to mitigate the effect of a mis-specified fit function in the real world is integrated with a modified version of the GFI, namely, the corrected GFI (CGFI). A simulation study is employed to compare the performance of improved RMSEA and GFI with the previously modified RMSEA and GFI, respectively. Furthermore, as there is no previous research considering the non-linear relationship between confounding variables, manifest variables, and latent variables, the proposed SSEM introduces nonparametric link functions, assumed to be monotone/non-monotone, to classical SEM instead of constructing a mixture multivariate model based on latent variables. Four types of regressions are used to predict the functional forms in the simulation study. Two metabolic syndrome datasets are used to demonstrate the application of the methods.
546 $a English
590 $a School code: 0076
650 4 $a Biostatistics. $3 1002712
650 4 $a Statistics. $3 517247
650 4 $a Mathematics. $3 515831
653 $a Categorical data
653 $a Dichotomous data
653 $a Fit index
653 $a Nonparametric function
653 $a Structural equation model
690 $a 0308
690 $a 0405
690 $a 0463
710 2 $a Georgetown University. $b Biostatistics. $e degree granting institution. $3 3816571
720 1 $a Makambi, Kepher $e degree supervisor.
720 1 $a Yuan, Ao $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 85-10B.
790 $a 0076
791 $a Ph.D.
792 $a 2024
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=31143197