View : 482 Download: 0

Full metadata record

DC Field Value Language
dc.contributor.author송윤수-
dc.creator송윤수-
dc.date.accessioned2016-08-26T02:08:22Z-
dc.date.available2016-08-26T02:08:22Z-
dc.date.issued1999-
dc.identifier.otherOAK-000000002182-
dc.identifier.urihttps://dspace.ewha.ac.kr/handle/2015.oak/193054-
dc.identifier.urihttp://dcollection.ewha.ac.kr/jsp/common/DcLoOrgPer.jsp?sItemId=000000002182-
dc.description.abstract본 논문에서는 관측자료가 우측절단자료 (right-censored data)인 경우에 회귀계수의 Rank-estimator를 구하기 위한 score function을 자료로부터 효과적(adaptive)으로 선택하는 방법을 전개하였다. 위험율 함수 (hazard function) h 와 그것의 도함수 h′를 구하기 위해 Mueller와 Wang (1990) 그리고 Uzunogullari 와 Wang (1992)이 제시한 locally adaptive smoothing technique 을 사용하였다. score function h′/ h 를 추정할 때 score function을 근사적으로 optimal하게 선택하는 방법에서 나타나는 약간의 결점을 보완하기 위해 log(h)를 비모수적으로 미분하는 대안을 제시하였다. 이 때 Friedman과 Stuetzle (1981)이 사용한 PPR이라는 Subroutine을 사용하였다. ; In this paper, we develop the adaptive choice of asymptotically efficient score functions for rank estimators of regression parameters in a linear regression model with right-censored data. Mueller and Wang(1990) and Uzunogullari and Wang(1992) provide the locally adaptive smoothing technique to estimate the hazard function h and its derivative h′from right-censored data. However, when score function h′/ h is computed for the asymptotically optimal choice of score functions, the naive estimator, which is just a ratio of estimated h′and h , is proven to have a few weaknesses. An alternative method which conquers the drawbacks of naive estimator is developed. In particular, a subroutine of the PPR(Projection Pursuit Regression) method coded by Friedman and Stuetzle (1981) is used to find the nonparametric derivative of log(h) for the problem of estimating h′/ h.-
dc.description.tableofcontentsABSTRACT 1. Introduction = 1 2. Locally Adaptive Hazard Smoothing with Right Censored Data = 4 2.1 Linear Rank Statistics = 4 2.2 Computation of Hazard Function = 7 2.3 Simulation Experiment = 12 3. Efficient Score Function for Rank Estimator = 16 3.1 Adaptive Rank Estimator = 16 3.2 Simulation Experiment = 18 4. Conclusion = 22 REFERENCES = 23 논문초록 = 25 감사의 글 = 26-
dc.formatapplication/pdf-
dc.format.extent703920 bytes-
dc.languagekor-
dc.publisher이화여자대학교 대학원-
dc.titleAdaptive rank estimator for right-censored data-
dc.typeMaster's Thesis-
dc.identifier.thesisdegreeMaster-
dc.identifier.major대학원 통계학과-
dc.date.awarded1999. 2-
Appears in Collections:
일반대학원 > 통계학과 > Theses_Master
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE