Interval estimation of quantile difference in the two-parameter exponential distributions

Abstract

We consider the interval estimation of the difference between the quantiles of two random variables with independent two-parameter exponential distributions based on Type II censored data. We derive asymptotic intervals based on the likelihood function, Bayesian intervals, as well as intervals based on the generalized pivot variable. We include some bootstrap intervals in our comparisons. The performance of the intervals is investigated in terms of their coverage probabilities and expected lengths using simulation techniques. An illustrative example is given.