Dear all,
I should start by same I'm trained as a biologist rather than a statistician, so am not fully grounded in statistical methods.
Basically I'm trying to find out how to specify the prior probability of a variance, given that its error is Chi square distributed. I'm using GenStat. For normally distributed variables GenStat has a function, llnormal, which allows log likelihoods to be calculated for a specified mean and variance (llnormal is described as 'log likelihood function for the Normal distribution'). There doesn't appear to be an equivalent for the Chi square distribution (no 'llchisquare') in GenStat. Can anyone help me on how to produce the same function for the Chi square? The nearest I can find is CLCHISQUARE, which calculates the cumulative lower probability for the Chi square distribution.
All the best
Richard.
___________________________________________________________
Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
now.
http://uk.answers.yahoo.com/
|