Combining multiple chains

[Filippo Ferrari]

Hi, 

I'm wondering what is the best practice after running multiple chains and checking for the Gelman-Rubin statistics as in the first basic HDDM tutorial.

Should parameters and point estimates be taken from just one of the chains (pick the chain associated with the model with the lowest DIC score), or is there a way to combine the different chains together? 

Thanks, 

Filippo 

[Michael J Frank]

Hi Filippo, yes you can combine chains together - as long as you've first checked they;ve converged. This is done automatically in HSSM when you plot posteriors via Arviz. In HDDM you can do that, you should see several threads on the archive of this list - just search "concat_models"

M 

--
You received this message because you are subscribed to the Google Groups "hddm-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hddm-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hddm-users/45ce1c5b-99f9-4cd3-83d6-3a73943a22b1n%40googlegroups.com.