Minimax estimators that shift towards a hypersphere for location vectors of spherically symmetric distributions (Q1072272): Difference between revisions

Let X be a p-dimensional random vector with density \(f(\| X-\theta \|)\) where \(\theta\) is a p-dimensional unknown location vector and \(''\| \| ''\) denotes the Euclidean norm. The author considers the minimax estimation of \(\theta\) by \({\hat \theta}\)(X) for \(p\geq 3\) with respect to a nondecreasing concave loss function \(\ell (\| {\hat \theta}-\theta \|^ 2).\) Minimax estimators \({\hat \theta}\)(X) are given (subject to certain conditions on f) which satisfy \(\| X\| \cdot \| {\hat \theta}(x)-X\| \leq C.\) If \(\theta\) is in a ball in \(R^ p\), then minimax estimators are given which shrink towards a hypersphere.