Generalized Wilcoxon Statistics for the Two-Sample Problem with Censored Data
Robert B. Latta · Biometrika · 1977
Several generalizations of the Wilcoxon statistic to the two–sample problem with censored data have been proposed. In this paper it is shown that Peto & Peto's generalization is equivalent to a modification of one that Efron proposed, in effect using weighted ranks. This equivalence suggests that Peto & Peto's test should be preferred to Gehan's test, which uses unweighted ranks.
Cited by 21 works. doi:10.1093/biomet/64.3.633