Comparing models through simulations

[VT]

Hi everybody, 

As many people here know, I ran Cafe5 using several models: a single lambda for the whole tree and more than one lambda model.

 My question is if there is a certain number of simulations of each model that would be considered "good" to do to perform a distribution.

I saw many people here asking similar questions about how to perform the comparison between models, but after all, I still don't have a very clear idea. I am sorry to bother everybody with this again. Any suggestions would be welcome.

Thanks in advance

V

[Hahn, Matthew]

1000 would be my recommendation, but 100 is likely fine.

cheers,

matt

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hahnlabcafe/ab295f51-7d05-4731-8317-df22918abb72n%40googlegroups.com.