Hello, everybody,
I am a newcomer here. just fell into the markov chain monte carlo area one
month ago. Basically our research is about inference algorithm using
particle filter for Bayesian network. we want to use Gibbs sampling or
metropolis-hasting method to move particles. The problem puzzled me is,
say, we know the full conditional distribution of one component random
variable is the conditional distribution given its markov blanket in the
Bayesian model, which is proportional to the product of the conditional
distribution of this variable given its parents and the conditional
distribution of each child of this variable given their parents. But we
don't know the normalizing constant usually.
So, quesion is: in case we do not have an analytical close form of the full
condtional distribution, how can we sample from the full conditional
distribution suppose we only know the full conditional distribution is
proportional to the product of some densities?
We are eager to have some hints or suggestions or explanation of some gibbs
sampling method to make our research further. Any reply about the above
question will be highly appreciated.
Best wishes to all,
Wei Sun
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|