Whoops, I forgot to add, many thanks to all who replied publicly and privately. I am very appreciative of all comments and suggestions. Mark Leeds was especially kind in terms of clarifying why my description of the situation was so confusing.
Cheers, Michael On Aug 26, 2006, at 2:51 PM, Michael Koppelman wrote: > Success! The line I needed was: > > gcoeffs <-nls(y~(a/b)*exp(-(x-c)^2/(2*b^2)),start=list > (a=0.4,b=2,c=-10), trace=TRUE) > > I also needed to provide good guesses for a, b and c. The attached > PNG should explain what I was going after, which is the line in the > center of that curve. I am sorry for my confusing and perhaps > incorrect usage of terminology. > > Cheers, > Michael ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.