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

Goodness-of-fit tests for density ra...

Deng, Xin.

FindBook

Google Book

Amazon

博客來

Goodness-of-fit tests for density ratio models.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Goodness-of-fit tests for density ratio models./
作者:	Deng, Xin.
面頁冊數:	95 p.
附註:	Adviser: Biao Zhang.
Contained By:	Dissertation Abstracts International67-12B.
標題:	Statistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3244430

Goodness-of-fit tests for density ratio models.
Deng, Xin.

Goodness-of-fit tests for density ratio models. - 95 p.

Adviser: Biao Zhang.

Thesis (Ph.D.)--The University of Toledo, 2006.

Several test statistics have been proposed for the purpose of assessing the goodness-f-fit tests for the logistic regression model based on case-control data. Zhang (1999) introduced a chi-square-type statistic to test the validity of the model, partitioning the data into several intervals by fixed pre-determined cutpoints. Enlightened by the goodness-of-fit tests of Moore (1975) and Hosmer and Lemshow (1982), we propose an alternative goodness-of-fit statistic by applying a general non-central chi-square test with several intervals, whose endpoints are calculated from a function of the data set. This data-driven test statistic with random partitions does not require a uniform rule to seek suitable fixed cutpoints case by case. Simulation studies show that powers of the proposed tests are better than those of the tests with the fixed partition. An analysis with respect to the number of groups is also presented in our study.Subjects--Topical Terms:

517247
Statistics.

Goodness-of-fit tests for density ratio models.
LDR:02352nam 2200265 a 45 001 963931
005 20110831
008 110831s2006 ||||||||||||||||| ||eng d
035 $a (UMI)AAI3244430
035 $a AAI3244430
040 $a UMI $c UMI
100 1 $a Deng, Xin. $3 1286989
245 1 0 $a Goodness-of-fit tests for density ratio models.
300 $a 95 p.
500 $a Adviser: Biao Zhang.
500 $a Source: Dissertation Abstracts International, Volume: 67-12, Section: B, page: 7152.
502 $a Thesis (Ph.D.)--The University of Toledo, 2006.
520 $a Several test statistics have been proposed for the purpose of assessing the goodness-f-fit tests for the logistic regression model based on case-control data. Zhang (1999) introduced a chi-square-type statistic to test the validity of the model, partitioning the data into several intervals by fixed pre-determined cutpoints. Enlightened by the goodness-of-fit tests of Moore (1975) and Hosmer and Lemshow (1982), we propose an alternative goodness-of-fit statistic by applying a general non-central chi-square test with several intervals, whose endpoints are calculated from a function of the data set. This data-driven test statistic with random partitions does not require a uniform rule to seek suitable fixed cutpoints case by case. Simulation studies show that powers of the proposed tests are better than those of the tests with the fixed partition. An analysis with respect to the number of groups is also presented in our study.
520 $a Also, we present methods to assess the adequacy of the stratified logistic regression model based on case-control data. Arbogast and Lin (2005) proposed methods which are derived from the cumulative sum of residuals over the covariate or linear predictor, which is mainly used and effective in the prospective approach. We propose a Kolmogorov-Smirnov-type statistic, which converges weakly to a zero-mean Gaussian process by the semi-parametric approach under the stratified density ratio model. The performance of the proposed methods are reported by simulations and some applications with the real data.
590 $a School code: 0232.
650 4 $a Statistics. $3 517247
690 $a 0463
710 2 $a The University of Toledo. $3 1023352
773 0 $t Dissertation Abstracts International $g 67-12B.
790 $a 0232
790 1 0 $a Zhang, Biao, $e advisor
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3244430