在本研究中,我們會提出一個含有時間相依共變數的擴充風險模型,使半母數存活分析更具調適性,此擴充風險模型包含了最常使用的Cox模型及加速失敗模型. 我們會利用計數過程及鞅論發展估計方程式來求取迴歸參數.在正規條件之下我們亦會研究參數估計的大樣本性質.在擴充風險模型的巢狀結構之下,我們可檢定Cox模型及加速失敗模型對資料之適合度.相關研究結果將應用於史丹佛心臟移植及台灣愛滋病資料上. ; In this study we propose a natural extension of Cox and AFT model, termed ”extended hazards (EH) model” for the analysis of survival data with time-depended covariates. The EH model refers to a general class of semiparametric regression models which encompasses both the Cox model and the accelerated failure time (AFT) model as its subclasses. A class of estimating equation using counting process and martingale theory is developed for estimating the regression parameters of the proposed model. The resulting estimators are shown to be consistent and asymptotically normal under appropriate regularity conditions. Since the EH model includes both the Cox model and the AFT model as its special cases such that we can perform tests by this nested structure to test the Cox and the AFT hypotheses. The simulation studies are carried out to evaluate the performance of the proposed model. The proposed approach will be applied to analyze the well-known Stanford heart transplant data and Taiwan AIDS data. ; 研究期間 9808 ~ 9907
