Bayes linear adjustment for variance matrices
Abstract
We examine the problem of covariance belief revision using a geometric approach. We exhibit an inner-product space where covariance matrices live naturally --- a space of random real symmetric matrices. The inner-product on this space captures aspects of our beliefs about the relationship between covariance matrices of interest to us, providing a structure rich enough for us to adjust beliefs about unknown matrices in the light of data such as sample covariance matrices, exploiting second-order exchangeability specifications.
Keywords
Cite
@article{arxiv.bayes-an/9506001,
title = {Bayes linear adjustment for variance matrices},
author = {Darren J Wilkinson and Michael Goldstein},
journal= {arXiv preprint arXiv:bayes-an/9506001},
year = {2008}
}
Comments
To appear in the Bayesian Statistics 5 conference volume. LaTeX, 11 pages, Chicago BIB-style (included), 2 postscript figures. Also available as a postscript file from http://fourier.dur.ac.uk:8000/~dma3djw/djwgvar.html For information on [B/D], go to http://fourier.dur.ac.uk:8000/stats/bd/