[Rd] extractAIC.nls

[Spencer Graves]

On 7/14/2012 6:39 PM, Gabor Grothendieck wrote:
> On Sat, Jul 14, 2012 at 8:48 PM, Spencer Graves
> <spencer.graves at prodsyse.com> wrote:
>> On 7/14/2012 4:11 PM, Gabor Grothendieck wrote:
>>> On Sat, Jul 14, 2012 at 6:53 PM, Spencer Graves
>>> <spencer.graves at prodsyse.com> wrote:
>>>> Hello:
>>>>
>>>>
>>>>         Does "extractAIC.nls" exist anywhere?  If no, is there some reason
>>>> it
>>>> should not be used?
>>>>
>>>>
>>>>         I'm planning to add it to the fda package, but before I do, I felt
>>>> a
>>>> need to ask if there is some reason I shouldn't or if there are caveats I
>>>> should include in the help page for it.
>>>>
>>> AIC in stats works on nls objects. It calls AIC.default which calls
>>> logLik.nls .
>>
>>
>>        Thanks for the reply.  Unfortunately, it fails for me in the following
>> modification of the first example in help('nls'):
> You tried it with extractAIC but my post suggested you use AIC which
> is a different function.

Hi, Gabor:  Thanks very much.  I now see that the help page for AIC 
refers to extractAIC and vice versa.  Spencer