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 [ author_sort:"liu, xin." ]

切換: 標籤 | MARC模式 | ISBD

A Bayesian nonparametric approach to...

Liu, Xin.

FindBook

Google Book

Amazon

博客來

A Bayesian nonparametric approach to testing essential unidimensionality in item response theory.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A Bayesian nonparametric approach to testing essential unidimensionality in item response theory./
作者:	Liu, Xin.
面頁冊數:	133 p.
附註:	Source: Dissertation Abstracts International, Volume: 68-01, Section: B, page: 0667.
Contained By:	Dissertation Abstracts International68-01B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3248861

A Bayesian nonparametric approach to testing essential unidimensionality in item response theory.
Liu, Xin.

A Bayesian nonparametric approach to testing essential unidimensionality in item response theory. - 133 p.

Source: Dissertation Abstracts International, Volume: 68-01, Section: B, page: 0667.

Thesis (Ph.D.)--University of Illinois at Chicago, 2006.

Bayesian nonparametric method estimates the true sampling density based on a nonparametric prior that gives support to all possible forms of densities. The Essential Unidimensionality model implies a strict subset of densities characterized with zero conditional inter-item covariance.Subjects--Topical Terms:

517247
Statistics.

A Bayesian nonparametric approach to testing essential unidimensionality in item response theory.
LDR:02084nmm 2200289 4500 001 1834478
005 20071119145700.5
008 130610s2006 eng d
035 $a (UMI)AAI3248861
035 $a AAI3248861
040 $a UMI $c UMI
100 1 $a Liu, Xin. $3 1282495
245 1 2 $a A Bayesian nonparametric approach to testing essential unidimensionality in item response theory.
300 $a 133 p.
500 $a Source: Dissertation Abstracts International, Volume: 68-01, Section: B, page: 0667.
500 $a Adviser: George Karabatsos.
502 $a Thesis (Ph.D.)--University of Illinois at Chicago, 2006.
520 $a Bayesian nonparametric method estimates the true sampling density based on a nonparametric prior that gives support to all possible forms of densities. The Essential Unidimensionality model implies a strict subset of densities characterized with zero conditional inter-item covariance.
520 $a The logarithmic distance of the set of the EU densities from the point estimate of the true density is measured by the Kullback-Leibler Divergence in terms of posterior expected log-likelihood ratio. The length of the distance indicates weight of evidence against the EU model, which defines the practical significance of evidence in terms of information units, and avoids the many fundamental statistical issues that are associated with p-values.
520 $a The simulation study showed that this approach displayed a moderate power in discriminating moderate multidimensionality from essential unidimensionality, and a high power in discriminating strong multidimensionality from moderate multidimensionality. The real illustrations of this approach in four educational and psychological test data implied that the proposed approach can be easily applied to various real test data.
590 $a School code: 0799.
650 4 $a Statistics. $3 517247
650 4 $a Psychology, Psychometrics. $3 1017742
690 $a 0463
690 $a 0632
710 2 0 $a University of Illinois at Chicago. $3 1020478
773 0 $t Dissertation Abstracts International $g 68-01B.
790 1 0 $a Karabatsos, George, $e advisor
790 $a 0799
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3248861