Hi! I’m trying to do a model comparison in a hierarchical meta model using a dummy variable that encodes model choice. Sampling from the full model doesn’t work well because the chains get me stuck in local optima: every chain “picks” one model and sticks to it, updating only the corresponding para…

This approach is taken in Kruschke’s DBDA book 2nd ed. Chapter 10. There’s a PyMC port by @cluhmann here https://nbviewer.org/github/cluhmann/DBDA-python/blob/master/Notebooks/Chapter%2010.ipynb . Maybe thus will be useful?

How to do model comparison with a dummy variable

Questions v5

chfin February 6, 2024, 12:06pm 9

I see. Thanks for the feedback!

Topic		Replies	Views
Mixture of hierarchical model Questions	8	3079	May 31, 2018
How to (re-)sample from an observed random (mixture) variable Questions	5	1667	April 9, 2018
Mixture of Linear models Questions	3	1206	July 21, 2018
Improve Sampling of Multilevel Regressions Model Questions	13	1270	March 9, 2020
Questions about the convergence of sampling the Mixture of hierarchical model Questions	2	1009	November 9, 2018

How to do model comparison with a dummy variable

Related topics