As an illustration, consider inference about the mean of an exponential distribution with density
The likelihood of a random sample is
(15) |
in terms of the sufficient statistic3
(16) |
Take as prior a density of the form
then the corresponding posterior is
(20) | |||
(21) |
which is clearly also of the form (17) and so may be written as
(22) |
(23) |
Hence (14) is closed under sampling, with (17) as a convenient natural conjugate prior family for inference about the mean . In such a case a complete analytic solution is available.
danny 2009-07-23