存活資料的關聯性分析 - 文獻回顧

Abstract

關聯性分析在許多應用研究中扮演關鍵角色。在此論文中，我們針對存活分析資料，回顧探討關聯性分析的重要文獻。我們先介紹描述整體與局部的關聯性測度，與應用中常被使用的copula模型。為因應不完整的存活資料，我們探討了統計推論的幾個可修正設限所帶來的偏誤的重要技巧。此外我們亦介紹更複雜的資料結構，藉以回顧其他的推論手法。最後我們進行了簡化的模擬分析。

Association analysis plays an important role in real-world applications. In this thesis, we review important literature on association study for survival data. First, we introduce commonly-seen global and local association measures and copula models which have been very popular due to their flexibility. Then we discuss statistical inference of these measures when data are subject to censoring. In particular we summary useful techniques which can correct the sampling bias due to censoring. We also other inference procedures which can handle complicated data structures. For illustrative purposes, we conduct some simulations to examine the finite-sample performance of different estimators.