Dear PyMC3 developers: Not sure what’s going on with pm.Multinomial() distribution, or it is my coding mistake as I’m new to PyMC3. I was trying to fit a Bayesian Hierarchical Model with the Yogurt data - starting with 2 households to 5 households. As the number of households increase from 2 to 5, …

Likely Metropolis is being very inefficient exploring the high dimensional space. Also, since you have n=1, you can use Categorical instead of Multinomial. Small rewrite of the model that seems to work: with pm.Model() as Yogurt_Model: # Specify prior distribution for population mean of alpha …

Problem with pymc3.Multinomial() distribution?

Questions

yanlingzuo July 14, 2020, 1:19pm 3

@junpenglao Thank you very much for suggesting using Categorical! Will give it try for the entire 100 household case. Thank you!

Topic		Replies	Views
Multinomial distribution doesn't behave as expected Questions	2	609	March 28, 2019
Why does the multinomial distribution work in this case? Questions	1	373	March 30, 2020
Posterior to Prior in Categorical distribution (or encapsulating multiple data sources to categorical analysis) Questions modeling	4	1812	February 28, 2022
Multinomial hierarchical regression with multiple observations per group ("Bad energy issue") Questions	22	3450	March 14, 2019
Multivariate categorical with different probabilities Questions	5	1890	April 9, 2019

Problem with pymc3.Multinomial() distribution?

Related topics