Estimation With LTRC Data And Doubly-censored Data Highlighting The Differences And Similarities

• Main Authors: Hsu,Min-yao, 續敏耀
• Other Authors: Shen,Pao-sheng
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/53661730720130708000

碩士 === 東海大學 === 統計學系 === 100 === Survival data are very common in many elds, e.g. medical science, demo- graphic, social science, and astronomy. The most typical characteristic of survival data is incomplete, where by far the most common models are those of censoring and truncation. Left-truncated right-censored (LTRC) often arise in epidemiology and individual follow-up studies. Their importance stems from the common use of prevalent cohort study designs to estimate survival from onset of a specied disease. The other types of data, called doubly-censored data, stem from occurrence of both left-censoring and right censoring in follow-up studies. The goal of this article is to highlight the dierences and similarities between the two types of data in a way that can help explaining some properties of the existing univariate and nonparametric bivariate estimators in literatures. Specically, for Cox model with both types of data, covariate, we demonstrate the dierence between partial-likelihood approach and full-likelihood approach.