Please join us for the Statistics Seminar scheduled from 11:30 a.m. to 12:30 p.m. on Friday, November 15. (in-person)

Speaker: Professor Siamak Noorbaloochi, University of Minnesota

Title: Bayesian Construction of Unbiased Estimators and Vice Versa

Abstract:

The fact that a non-trivial estimator cannot be both unbiased and the posterior mean has been used against unbiasedness property. I will argue that indeed these two concepts are complementary: bias is a measure of goodness of the Bayes estimator.  Unbiasedness and being Bayes (against squared error loss) are two adjoint linear operators. Gibbs sampling is a possible tool to construct unbiased estimators. Bayes estimators are constructible from a corresponding unbiased estimator.

Zoom information:

Meeting ID: 979 6542 3067

Passcode: 48464114