[ADMB Users] a couple of questions about MCMC

William Stockhausen William.Stockhausen at noaa.gov
Thu Oct 13 18:48:13 PDT 2011


Hi all,

Couple of questions about mcmc in ADMB:

1.  What does "resuming" an mcmc run using the command line flag "-mcr" 
mean?  Can I use -mcr to resume an interrupted mcmc run from the point 
it was interrupted (by, say, a computer crash or terminating the DOS 
window)?  Or does -mcr only work with a successfully-completed mcmc run?

2.  In the mceval phase following an mcmc run, I calculate a number of 
values in the procedure section and write them to a file. Does the first 
set of values (presumably the first time through the procedure section 
during the mceval phase) that gets written out correspond to the set of 
values I'd get using the model parameters that minimize the objective 
function (maximize the likelihood)?  A quick comparison of the std file 
values and the mceval results suggests this is the case (and it'd be 
mighty convenient if it were), but I'd like to confirm it.

Thanks for any and all help!

Buck

-- 
***************************************************
* Dr. William T. Stockhausen                      *
*_________________________________________________*
* Resource Ecology and Fisheries Management       *
* Alaska Fisheries Science Center                 *
* National Marine Fisheries Service               *
* National Oceanic and Atmospheric Administration *
* 7600 Sand Point Way N.E.                        *
* Seattle, Washington 98115-6349                  *
*_________________________________________________*
* email: William.Stockhausen at noaa.gov             *
* voice: 206-526-4241 fax: 206-526-6723           *
* web  : http://www.afsc.noaa.gov                 *
***************************************************
All models are wrong, some are useful.--G.E.P. Box
Beware of geeks bearing equations.    --W. Buffett
***************************************************
Disclaimer: The opinions expressed above are personal
and do not necessarily reflect official NOAA policy.




More information about the Users mailing list