On rates of uniform convergence of lower extreme generalized order statistics (Q2488460)

A family of generalized order statistics is considered which includes ordinary order statistics and progressive type II censored order statistics. It is shown that under a von Mises-type condition the joint distribution of suitably normalized lower order extreme generalized order statistics converges w.r.t. the variational distance to the same limit for ordinary order statistics. For distributions in a \(\delta\)-neighborhood of a generalized Pareto distribution, the rate of convergence is \(\sim n^{-\min(\delta,1)}\). These results are applied to derive a normal approximation for the distribution of an estimate of the Weibull-type distribution shape parameter.