摘要 本文針對具右設限的成對存活資料，提出調整成對資料相關性的成對對數秩近似檢定及成對限制平均壽命差異的近似檢定。除以實例說明其應用，並且採用模擬方法研究其型I誤差率及檢定力表現。本文研究具右設限的相關成對存活資料中兩個處理效應的比較。我們提出經過相關性調整後的對數秩檢定與限制平均存活時間差異的檢定。本文除以一個實例說明檢定的應用，也以模擬方法研究所提檢定方法的型I誤差率及檢定力表現。 Abstract In this article, we are concerned with the problem of comparing two treatment effects when correlated paired survival data are subjected to random right-censorship. The logrank test and the test based on the difference of two restricted mean survival time adjusted by the survived correlation are proposed. A real dataset is illustrated for the use of the tests and a simulation study is conducted to investigate the Type-I error and power performances of the proposed tests.
