Michael: Your parameter specification is probably over-determined, so that you have an infinite set of parameters that give essentially the same solution within numerical error. I would venture to guess that this will not be fixable with alternative optimizers. It is up to you to provide a sensible problem specification; failure to do so cannot be blamed on the optimizer.
Cheers, Bert On Thu, Sep 5, 2013 at 3:23 AM, Michael Meyer <spyqqq...@yahoo.com> wrote: > Thanks for all replies. > The problem occurred in the following context: > > A Gaussian one dimensional mixture (number of constituents, locations, > variances all unknown) > is to be fitted to data (as starting value to or in lieu of mixtools). A > likelihood maximization is performed. > > I'll try to destill the code so that reproducible failure of L-BFGS-B > occurs > and post it here. > > > > Michael Meyer > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Bert Gunter Genentech Nonclinical Biostatistics Internal Contact Info: Phone: 467-7374 Website: http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
