Quantile estimation of treatment effect for the two-sample problem with right censored data

Estimating the treatment effect in the two-sample problem when some of the data are right censored is an important problem. Akritas (1986) and Meng et al. (1991) proposed estimators of the treatment effect under the location shift model using the fact that the median of the distribution of the difference of two independent random variables is the location shift. An estimator based on the quantile function is proposed and consistency and asymptotic normality of it are also presented.