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

On the hazard function with a change...

Cho, Yong Won.

FindBook

Google Book

Amazon

博客來

On the hazard function with a change point.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	On the hazard function with a change point./
作者:	Cho, Yong Won.
面頁冊數:	170 p.
附註:	Source: Dissertation Abstracts International, Volume: 50-01, Section: B, page: 0039.
Contained By:	Dissertation Abstracts International50-01B.
標題:	Biology, Biostatistics. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=8907543

On the hazard function with a change point.
Cho, Yong Won.

On the hazard function with a change point. - 170 p.

Source: Dissertation Abstracts International, Volume: 50-01, Section: B, page: 0039.

Thesis (Ph.D.)--University of California, Los Angeles, 1988.

This dissertation studied situations where survival distributions could be described through the use of two different failure models, the "early" and "late" failure rates, which were separated by a change-point parameter Subjects--Topical Terms:

1018416
Biology, Biostatistics.

On the hazard function with a change point.
LDR:02899nmm 2200289 4500 001 1865092
005 20041216142211.5
008 130614s1988 eng d
035 $a (UnM)AAI8907543
035 $a AAI8907543
040 $a UnM $c UnM
100 1 $a Cho, Yong Won. $3 1952555
245 1 0 $a On the hazard function with a change point.
300 $a 170 p.
500 $a Source: Dissertation Abstracts International, Volume: 50-01, Section: B, page: 0039.
500 $a Chair: William G. Cumberland.
502 $a Thesis (Ph.D.)--University of California, Los Angeles, 1988.
520 $a This dissertation studied situations where survival distributions could be described through the use of two different failure models, the "early" and "late" failure rates, which were separated by a change-point parameter $\ tau $. The discontinuous hazard function with a change-point, h(t) = a I(0 $\ leq $ t $\ leq \tau $) + b I(t $> \tau\rbrack $ for a $> $ b $> $ 0, where I(A) is the indicator function of the set A, was investigated and it was shown that the likelihood ratio test based on the large sample theory could not be applied for testing a null hypothesis regarding the change-point parameter. Two test statistics for testing hypotheses were proposed: one using a stepped likelihood ratio statistic and the other a smoothed likelihood ratio statistic. The stepped likelihood ratio statistic was obtained by replacing the log-likelihood function with a step function, and the smoothed likelihood ratio statistic from using an arctangent function as an approximation to the discontinuous hazard. The smoothed likelihood ratio test gave reasonable sizes for tests at various significance levels. The power of the smoothed likelihood ratio test was evaluated by simulation experiments, and a larger ratio, r = a/b, gave a more power.
520 $a When the assumed model was the arctangent hazard function, a simulation study to evaluate three tests for a simple hypothesis regarding $\ tau $ was done: the score test, the likelihood ratio test, and the test based on the asymptotic normality of the MLE. The score test gave reasonable sizes of test at all levels of factors, but other tests seemed to depend upon the ratio, r = a/b, and $\ beta $ for samples with small n. The power of the score test also depended upon the ratio; a larger ratio being associated with a higher power.
520 $a In applying the discontinuous hazard function to kidney transplant data, it was found that the change-point parameter for second cadaveric graft survival was highly dependent upon the selection period which was based on the date of transplantation.
590 $a School code: 0031.
650 4 $a Biology, Biostatistics. $3 1018416
650 4 $a Health Sciences, Medicine and Surgery. $3 1017756
690 $a 0308
690 $a 0564
710 2 0 $a University of California, Los Angeles. $3 626622
773 0 $t Dissertation Abstracts International $g 50-01B.
790 1 0 $a Cumberland, William G., $e advisor
790 $a 0031
791 $a Ph.D.
792 $a 1988
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=8907543