Hi, I want to calculate the WBIC, which requires multiplying the log-likelihood of the model with a constant. In Stan, it is very easy to do this. For PyMC, I can think of a way by defining a custom distribution with a modified log-likelihood for every distribution needed in the likelihood. This c…

The custom sampler is the right approach. Potential doesn’t allow for a temperature parameter that changes over draws

Best way to multiple a term to the log-likelihood

cluhmann March 11, 2024, 5:00pm 2

Welcome!

If you call pm.sample() with idata_kwargs={"log_likelihood": True}), then the log likelihoods are just sitting in the returned inferenceData object. Does that help?

1 Like

Topic		Replies	Views
How to set up a custom likelihood function for two variables Questions	8	13306	March 9, 2018
DensityDist in PyMC4 Development	2	1007	February 12, 2020
Likelihood evaluation and DensityDist Questions	2	1709	February 21, 2018
Likelihood given by product of two distributions? Questions	6	2215	August 4, 2018
How are multiple likelihoods handled in PyMC? Questions	1	1325	November 8, 2017

Best way to multiple a term to the log-likelihood

Related topics