95% HPD

[pete cowman]

hi all,

just wondering if some one could tell me if the 95% HPD around a node
height for a beast analysis is comparable to a 95% confidence
interval?

thanks

Pete

[Andrew Rambaut]

Dear Pete,

That kind of depends what you think confidence intervals mean:

http://en.wikipedia.org/wiki/Confidence_interval#Meaning_and_interpretation

HPDs are a form of credible interval (the Bayesian analog) but as with
most things Bayesian, they have a much more straight forward
interpretations:

"For example, a statement such as "following the experiment, a 90%
credible interval for the parameter t is 35-45" means that the
posterior probability that t lies in the interval from 35 to 45 is
0.9. In general, Bayesian credible intervals do not coincide with
frequentist confidence intervals as the credible interval incorporates
problem-specific contextual information from the prior distribution
whereas confidence intervals are based only on the data."

Quoted from:
http://en.wikipedia.org/wiki/Credible_interval

Andrew

On 6 Aug 2008, at 08:36, pete cowman wrote:

> just wondering if some one could tell me if the 95% HPD around a node
> height for a beast analysis is comparable to a 95% confidence
> interval?

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.