English

Local computation of influence propagation through Bayes linear belief networks

bayes-an 2008-02-03 v1 Data Analysis, Statistics and Probability

Abstract

In recent years there has been interest in the theory of local computation over probabilistic Bayesian graphical models. In this paper, local computation over Bayes linear belief networks is shown to be amenable to a similar approach. However, the linear structure offers many simplifications and advantages relative to more complex models, and these are examined with reference to some illustrative examples.

Keywords

Cite

@article{arxiv.bayes-an/9609001,
  title  = {Local computation of influence propagation through Bayes linear belief networks},
  author = {Darren J Wilkinson},
  journal= {arXiv preprint arXiv:bayes-an/9609001},
  year   = {2008}
}

Comments

LaTeX2e, 11 pages. In submission. See http://www.ncl.ac.uk/~ndjw1/ Also available from http://www.ncl.ac.uk/~ndjw1/pub/djwlc.html For more information about Bayes linear methods, see the Bayes linear home page: http://fourier.dur.ac.uk:8000/stats/bayeslin/