next up previous contents
Next: Priors Up: Analysis Previous: Analysis

Likelihood

The likelihood is just the binomial, with parameters tex2html_wrap_inline3081 . Let tex2html_wrap_inline3083 be the data, tex2html_wrap_inline3085 be the data up to time point N, and tex2html_wrap_inline3089 be the parameters, with tex2html_wrap_inline3091 being the hyperparameters. Let tex2html_wrap_inline2033 represent the collection of all tex2html_wrap_inline2725 .

Then, the likelihood is given as

equation1046

since tex2html_wrap_inline3097 independent of tex2html_wrap_inline2131 conditional on tex2html_wrap_inline2033 . And, this is just

equation1051

Certainly one is interested in predicting tex2html_wrap_inline3103 given tex2html_wrap_inline3105 and the tex2html_wrap_inline3107 .



Cathal Walsh
Sat Jan 22 17:09:53 GMT 2000