[R] nls NAs

[Elizabeth Webb]

Hello all-

I got some help on this.  While it did not fix the problem with nls, one can
get around the problem by creating a data frame with the NAs excluded.  As
follows:

goodidx<-which(! is.na(warming$T10cm))
warming2<-warming[goodidx,]
warm.10<-nls(umoles60~alpha*exp(beta*T10cm),start = start,data=warming2)

Note there is also a package on on CRAN(nlmrt) that tries to be more
aggressive in finding a
solution to nls type problems. The function wrapnls() in nlmrt runs a
Marquardt code then
nls (same calling sequence, but will not yet handle weights, nor every
possible
expression, though it does most).  

Elizabeth



--
View this message in context: http://r.789695.n4.nabble.com/nls-NAs-tp4644689p4646182.html
Sent from the R help mailing list archive at Nabble.com.