Estimation of bivariate survival function under random censoring이변량 중도절단자료의 생존함수 추정에 관한 연구

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 533
  • Download : 0
In this thesis when the bivariate censored data are given, it is considered an estimation of survival function and its applications - estimation of the treatment effect between two groups and testing problem of bivariate symmetry. There have been many literatures for analysis of bivariate censored data. For independent two-sample problems, it is analyzed separately by means of well understood univariate techniques. However, analysis of each marginal distribution ignores valuable information about the inter-relationships among two groups, and indeed can even lead to paradoxical results. Motivated by these facts, for dependent bivariate censored data, it is considered the problems of estimation of bivariate survival function, treatment effect and testing problem of bivariate symmetry, which are all important in engineering and biomedical science. For univariate censored data, the estimation problem of survival function has been studied by using well known Kaplan and Meier (1958) estimator. It can be easily applicable to the independent bivariate censored data. However, the survival function estimators, which have been considered for dependent bivariate censored samples, are complicated and most of these do not satisfy monotone property. In chapter 3, new estimators are considered under three kind of censoring schemes using adoption of Burke (1988)``s idea. The first censoring scheme is that one component can be completely observable. The second one is that two survival time can be censored by one censoring time - univariate censoring - which can be occurred very often when the study periods are ended but the patients or units are still alive. The third one is a bivariate censoring which is a direct generalization of one-sample cases. The asymptotic normalities of proposed estimators are shown and the asymptotic covariances are computed. Survival function estimators can be used for the estimation of treatment effect. Under the location-shift model assumption, ...
Advisors
Kim, Byung-Chunresearcher김병천researcher
Description
한국과학기술원 : 경영공학전공,
Publisher
한국과학기술원
Issue Date
2000
Identifier
158287/325007 / 000935207
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 경영공학전공, 2000.2, [ vi. 113 p. ]

Keywords

Estimation; Censored data; Bivariate symmetry; Bivariate survival function; Treatment effect; 처리효과; 추정; 중도절단자료; 이변량 대칭성; 이변량 생존함수

URI
http://hdl.handle.net/10203/53331
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=158287&flag=dissertation
Appears in Collection
KGSM-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0