在醫學研究中，經常受限於資料蒐集的方式，獲得具備右截斷特性的資料(right truncated data)。針對此類特性資料的存活函數之估計與檢定方法，學者Lagakos, Barraj和Gruttola(1988)建議將時間軸反轉(reverse time)，使右截斷資料轉成左截斷資料，再應用加權對數秩檢定(weighted logrank test)來判斷兩組存活函數是否相等。但此方法在權數的選取上，必須知道兩組資料的分配才有辦法得知其權數，故在使用上有所限制。因此，本論文將依據Efron與Petrosian(1999)所提出以涉險個數為主的演算法，不經由反轉時間軸便能直接估計出在每一時間點的涉險個數，進而應用加權對數秩檢定，進行兩存活函數是否相等之檢定。此外，本論文將以模擬方式比較上述兩種檢定方法之檢定力。

Right-truncated data arise naturally from some sampling schemes, such as cross-sectional sampling, and retrospective sampling, in many research areas. Failure to account for right truncation can lead to biased inference. Consequently, Lagakos et al. (1988) proposed weighted logrank test based on reverse times to compare the survival functions. Their approach may be straightforward for statisticians, but the implementation of their method may not be easy for practical researchers. Therefore, this paper proposes a log-rank type test based on estimated number of subjects in the risk set in forward time. The comparative results from a simulation study are presented and the implementation of these methods to the AIDS transfusion data is presented.

Bilker, W. B. and Wang, M. C. (1996). A Semiparametric Extension of the Mann-Whitney Test for Randomly Truncated Data. Biometrics 52, 10-20.

Efron, B. and Petrosian, V. (1999). Nonparametric Methods for Doubly Truncated Data. Journal of the American Statistical Association 94, 824-834.

Hyde, J. (1980). Survival Analysis with Incomplete Observations. In Biostatistics Casebool, R. G. Miller, B. Efron, B. W. Brown, and L. E. Moses, eds. New York: John Wiley and Sons, 31-46.

Lagakos, S. W. , Barraj, L. M. and Gruttola, V. DE (1988). Nonparametric Analysis of Truncated Survival Data, with application to AIDS. Biometrika 75, 515-523.

Wang, M. C. , Jewell, N. P. and Tsai, W. Y. (1986). Asymptotic Properties of the Product Limit Estimate under Random Truncation. The Annals of Statistics 14, 1597-1605.