[18866] in s-news-athena
Re: [S] nls() fails in a loop
daemon@ATHENA.MIT.EDU (Patrick Burns)
Thu Aug 5 10:29:54 1999
Date: Thu, 5 Aug 1999 10:20:34 -0400
From: pburns@epafs1.sbi.com (Patrick Burns)
Message-Id: <199908051420.KAA12848@epacs2.sbi.com>
To: Bill.Venables@cmis.CSIRO.AU, bates@stat.wisc.edu
Cc: mim78@cict.fr, s-news@wubios.wustl.edu
Doug Bates writes:
> <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.
There is a function in S Poetry called "ignore.error" which is a
general solution to escaping an error condition in cases where
there isn't an easy workaround. It uses "restart" as Bill
suggested.
http://www.seanet.com/~pburns/Spoetry
http://www.i-way.co.uk/~pat/Spoetry
Pat Burns
-----------------------------------------------------------------------
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
