Estimation for the Linear Model with Uncertain Covariance Matrices
Statistics Theory
2014-03-12 v1 Statistics Theory
Abstract
We derive a maximum a posteriori estimator for the linear observation model, where the signal and noise covariance matrices are both uncertain. The uncertainties are treated probabilistically by modeling the covariance matrices with prior inverse-Wishart distributions. The nonconvex problem of jointly estimating the signal of interest and the covariance matrices is tackled by a computationally efficient fixed-point iteration as well as an approximate variational Bayes solution. The statistical performance of estimators is compared numerically to state-of-the-art estimators from the literature and shown to perform favorably.
Cite
@article{arxiv.1401.7195,
title = {Estimation for the Linear Model with Uncertain Covariance Matrices},
author = {Dave Zachariah and Nafiseh Shariati and Mats Bengtsson and Magnus Jansson and Saikat Chatterjee},
journal= {arXiv preprint arXiv:1401.7195},
year = {2014}
}