[18852] in s-news-athena
Re: [S] nls() fails in a loop
daemon@ATHENA.MIT.EDU (Douglas Bates)
Wed Aug 4 11:58:39 1999
To: "Venables, Bill (CMIS, Cleveland)" <Bill.Venables@cmis.CSIRO.AU>
Cc: "'ROS Mathieu'" <mim78@cict.fr>, s-news@wubios.wustl.edu
From: Douglas Bates <bates@stat.wisc.edu>
Date: 04 Aug 1999 10:52:14 -0500
In-Reply-To: "Venables, Bill's message of "Wed, 4 Aug 1999 17:39:19 +1000"
Message-Id: <6r1zdjr1ap.fsf@verdi.stat.wisc.edu>
"Venables, Bill (CMIS, Cleveland)" <Bill.Venables@cmis.CSIRO.AU> writes:
> 3. As an absolute last resort, and with great care and attention, cautiously
> approach the restart() function. It is designed to step over failing
> calculations in exactly this sort of context, but it is not for the faint
> hearted. You can do a lot of damage with this one...
I agree with everything that Bill says, especially trying the
"plinear" algorithm. Before trying to use the restart function
yourself, you might check the nlsList function from the NLME library
and see if you can use that. It implements the approach that Bill
describes.
The NLME library, whose latest version was announced here recently, is
available at
http://nlme.stat.wisc.edu/
http://cm.bell-labs.com/stat/NLME
-----------------------------------------------------------------------
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
