Hi, I’m trying to use a posterior distribution as a likelihood (first in a new model but eventually I want to merge these steps into one model). Is there a way to handle the posterior from model_01 as a likelihood in model_02? My toy code looks like this: #simulating some data N = 1000 stimulus…

Using posterior as likelihood

ricardoV94 November 3, 2022, 10:49am 3

There is also a new utility in pymc_experimental that uses a MvNormal approximation IIRC: API Reference — pymc_experimental 0.0.1 documentation

4 Likes

Topic		Replies	Views
How to set up a custom likelihood function for two variables Questions	8	13306	March 9, 2018
Plot the likelihood instead of posterior distribution Questions	5	2652	May 19, 2019
Feed a posterior as an input random variable in another model Questions	1	342	October 6, 2020
Unexpected results Questions	2	507	February 15, 2019
Example of computing posterior from prior and likelihood under the hood version agnostic modeling	3	1504	May 11, 2022