TJ McKinley

Thanks Daniel. That makes sense, and your approach to model the conditionals aligns with what we were thinking of doing to overcome this in this case, so that's great! However,...

No worries at all. Thanks for your reply and thanks once again for developing NIMBLE.

So I've been thinking about this and I'm not quite following why analytical forms for the conditionals would be required for a RW M-H algorithm, since you could just use...

In case it's of interest, I've attached some code below which seems to work without requiring the conditionals. I've used the standard RW and RW_block sampler code, but amended to...

Hi Daniel, No worries at all and thanks for the response. Of course, there's no pressure to adopt any of these ideas, but if they're useful to build upon then...