[ADMB Users] although the objective function is not calculated, all parameters are estimated with even uncertainty

Saang-Yoon shyunuw at gmail.com
Wed May 18 20:53:59 PDT 2011

I have interesting results.   Although the "ultimate" objective
function becomes "nan" (i.e., f = nan), all point estimates of
parameters look very reasonable, and the correlation and standard
deviation outputs (e.g., xxx.cor, xxx.std) are also generated.   Do
you believe these results are reliable?

More information about the Users mailing list