[R-sig-ME] Extending chains in MCMCglmm?
alexandre.courtiol at gmail.com
Sun Mar 13 21:55:26 CET 2016
Is there a way to extend MCMC chains once a MCMCglmm fit is over?
I am using a relatively big dataset (35000 rows) and a rather complex model
(4 response variables, several fixed and random effects). So it is (as it
should be) slow.
Therefore after a run of one week, if the MCMC chains did not converge
well, I would like to know how I could continue to run the model starting
from the end point of the chains I have obtained. Ideally, I would love to
be able to consider the full length of the chains (so old + new ones) and
to have to possibility to redefine the burn-in period for the analyses.
Apologies if it is in the manual/course notes/tutorial... I looked for it
but may have missed it.
*"Science is the belief in the ignorance of experts"*, R. Feynman
[[alternative HTML version deleted]]
More information about the R-sig-mixed-models