We consider a superposition of an unknown number of independent homogeneous Poisson processes in which the source of each event can be identified. After observing the system for a fixed time t, the total rate, U(t), of the unobserved processes is to be estimated. We prove that a uniformly...

This paper presents Pitman closest equivariant estimators of the location of elliptically symmetric distributions. In the scalar parameter case the best estimator is the same for all loss functions that satisfy certain mild monotonicity conditions, which makes the estimator very useful when the...

Non-homogeneous Poisson Process (NHPP) models form a significant subclass of the many software reliability models proposed in the literature. We prove an important limitation of NHPP models for which the expected number of failures in infinite testing is finite. Specifically, the parameters of...

Some distributional results related to the recapture debugging experiment of Nayak (1988) for estimating the number of errors, N, in a software are obtained. It is proved that while the sufficient statistics are complete, an unbiased estimator of N based on them does not exist.