Summary: We show how to combine posterior probabilities from an emsemble of models, each of which estimates the same parameter (or class) but using "independent" data. From this we describe how to separate out and replace the class prior (or the model-based prior) of a classifier post hoc and how this relates to the combination problem ...

 

Bibtex entry for this abstract:

@UNPUBLISHED{LL:CBJ-053,
author = {C.A.L.~Bailer-Jones and K.~Smith},
title={{C}ombining probabilities},
institution={Max-Planck-Institute for Astronomy, Heidelberg},
year={2011},
month={July},
url={http://www.rssd.esa.int/doc_fetch.php?id=2968255},
note={GAIA-C8-TN-MPIA-CBJ-053},
type={Technical note}
}