[S] False Convergence with nlminb

Norm Phillips (nphillips@bccancer.bc.ca)
Mon, 22 Jun 1998 11:36:03 -0700


I am using Splus3.4 on a unix workstation. I used nlminb to
successfully fit 13 parameters with a negative log-likelihood, but
when I make the model more complex and add an additional
6 parameters, and after increasing the maximum number of
iterations and function evaluations, I get a FALSE CONVERGENCE
message. Does anybody have any suggestions as to what
might cause this and what I might do to remedy it?
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news