Estimation after subset selection from exponential populations

From MaRDI portal





Let \(\pi_ 1,...,\pi_ k\) be k independent exponential populations with means \(\theta_ 1,...,\theta_ k\), respectively. Let \(Y_ 1,...,Y_ k\) denote the sample means based on n independent observations from each population. For selecting a nonempty subset containing the best population (the one associated with the largest \(\theta_ i)\) we consider the rule of Gupta which selects \(\pi_ i\) if and only if \(Y_ i\geq c \max_{1\leq j\leq k}Y_ j\), where \(0<c<1\) is determined such that the probability of including the best population in the selected subset is at least \(P^*\), a preassigned level. We consider the problem of estimating M, the mean of all \(\theta_ i's\) in the selected subset, for \(k=2\). It is shown that the natural estimator T is positively biased. An adjustment for bias of T is considered so that this adjusted estimator \(T_{\lambda 1}\) is unbiased when \(\theta_ 1=\theta_ 2\). The performances of T and \(T_{\lambda 1}\) are compared in terms of bias and mean square error for selected values of \(P^*\), n, and \(\rho =\theta_ 1/(\theta_ 1+\theta_ 2)\).




Cited in
(26)








This page was built for publication: Estimation after subset selection from exponential populations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3736704)