[Esip-preserve] Uncertainty, Hypothesis Testing, and Related Issues

[Bruce Barkstrom via Esip-preserve]

The ISO standards that provide Guidance on the Statement of Precision
in Measurements (where the authority is the International Bureau of
Weights and Measures) appears to be moving in the direction of using
Bayesian statistics, rather than the classical "frequentist" approach.
Berger, J. O., 1985: Statistical Decision Theory and Bayesian Analysis,
2nd ed., Springer, New York, NY makes a very strong case for using
Bayesian analysis.

In section 4.3.3 on Hypothesis Testing [pp. 145-157], he compares
the Bayesian approach with the usual hypothesis testing one.  The
results of the analysis could be markedly different.  His examples
suggest that discussions of repeatability and sampling as well as
hypothesis testing could be strongly influenced by the analysis methods.
I don't know the current sense of the bioinformatics community on this
difference in approach, but it does seem to warrant some discussion.

Bruce B.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.esipfed.org/pipermail/esip-preserve/attachments/20140805/e042a840/attachment.html>