A Generalized Gehan and Gilbert Test for Paired Observations that are Subject to Arbitrary Right Censorship
提出一种渐近分布自由的检验方法，用于处理任意右删失的配对观测数据，在双变量指数模型下比符号检验更有效。
Abstract An asymptotically distribution-free test, along the line of Gehan (1965) and Gilbert (1962), for paired observations that are subject to arbitrary right censorship is proposed. This test is shown to be more powerful than the sign test under a bivariate exponential model.