Bayes linear methods home page

 

Bayes linear methods are subjective statistical analyses based on expectation and covariance structures, rather than on distributional assumptions.


Bayes linear publications

The Bayes linear methods series of technical reports provides an introduction to Bayes linear methods, and a tutorial on solving basic problems using [B/D] (the Bayes linear computer programming language).

The book Bayes linear statistics: theory and methods, by Michael Goldstein and David Wooff, was published by Wiley in April 2007. ISBN 978-0-470-01562-9. Errata can be found here.

Michael Goldstein has written the article Bayes Linear Analysis, included in the Encyclopaedia of Statistical Sciences (update vol 3, 1998, Wiley, reproduced by permission), which provides an overview of the approach.

Statisticians currently working on or developing Bayes linear methodology include:

 Peter Craig (University of Durham)

 Frank Coolen (University of Durham)

 Malcolm Farrow (University of Newcastle)

 Kostas Triantafyllopoulos (University of Sheffield)

 Michael Goldstein (University of Durham)

 Tony O'Hagan (University of Sheffield)

 Allan Seheult (University of Durham)

 Simon Shaw (University of Bath)

 Darren Wilkinson (University of Newcastle)

 David Wooff (University of Durham)

 

Bayes linear computing

[B/D] - The Bayes linear computer programming language
BAYES-LIN - Object-oriented software for Bayes linear local computation