This note uses the method of Wilks [2, 0 11.51 to construct non-parametric tolerance limits when sampling has Type I censoring, as defined in 0 1O.C.l of [l]. A sample of n items is tested. The result of testing the p-th item is the continuous random variable X, , which can be observed only in case it exceeds a number T. X, = {X, : 1 _ T. We require a tolerance limit of level ((Y, 0) for G, i.e., we require to define a random variable X, = X,,(J’,) such that if G is a distribution function, then
No takes yet. Share an insight, caveat, or question.
Robert Bohrer (1968) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: