On a admissible estimation for parametric functions in linear models

From MaRDI portal





Consider a linear model E y\(=\theta \in K\), cov \(y\in V\), and let (a,y) be a linear estimator of some parametric function (p,\(\theta)\). Under the squared error loss function \(((a,y)-(p,\theta))^ 2\) the authors investigate whether or not (a,y) is admissible for (p,\(\theta)\). They also discuss when a linear mapping C: \(K\to K\) has the property that (a,Cy) is an admissible estimator of (Ey,a) for all \(a\in K\). The question of replacing inadmissible estimators by admissible ones is also discussed.











This page was built for publication: On a admissible estimation for parametric functions in linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3823619)