how can i improve effective size of log likelihood in a phenotypic rates analysis?

22 views

Skip to first unread message

Jamie T

unread,

Jul 26, 2023, 2:02:14 PM7/26/23

to bamm-project

hello,

i am running phenotypic rates on a large tree (>10,000 taxa), and did not expect either N_shifts or loglik to converge very well. N_shifts has arrived at an ESS of >200, but loglik is 2.3. Is there any way to change the weight of any parameter to improve convergence?