[R-sig-ME] Help! What are the typical diagnosis that one can do when, facing

[Ben Bolker]

On 12-05-10 02:05 PM, Joshua Wiley wrote:
> On Thu, May 10, 2012 at 10:46 AM, Ben Bolker <bbolker at gmail.com>
> wrote:
>> Highland Statistics Ltd <highstat at ...> writes:
>> 
>> 
>>>> Help! What are the typical diagnosis that one can do when
>>>> facing "fail to converge"? Could anybody please shed some
>>>> lights on us, besides increasing the number of iteration
>>>> limit?
>> 
>> [snip]
>> 
>>> 1. Simplify your model 2. Center or standardise your
>>> covariates 3. Increase the number of iterations 4. Lower the
>>> convergence criteria 5. Provide starting values.
>> 
>> Very nice.
>> 
>> #6 try an alternative package/optimization platform, if one is
>> available (in this case, if you are using a complex model for
>> heteroscedasticity and want to stay within R, probably not ...)
> 
> You could specify a different optimization method via the control 
> argument.

 [snip]

 Thanks, I had forgotten about this (or didn't know).  See ?lmeControl
and particularly the "opt" and "optimMethod" arguments: you can choose
"L-BFGS-B".  I don't know what happens if you (1) specify a
non-derivative-based method such as Nelder-Mead; (2) specify a method
that doesn't allow box constraints ...