Hi all,
I'm trying to calculate a nonlinear regression to get a simpler expression
for a complicated formula.
I expect that a function of the type
1
----------------------------
- (a + b1*x + b2*y)
1 + e
should be a good fit. Now I'm trying to calculate a regression with R and
must admit that I'm slightly confused by the variety of possibilities to run
this calculation. My first approach was to calculate the regression with the
least-squares-method and "optim" manually. This seems to work quite
well.
"An Introduction to R" points out that this can be done with
"nlm" which
seems to work for me too. Now "nls" and "gnlm" seem to point
into this
direction as well. Could anyone advise me, if I'm on the right track and
what the advantages and disadvantages of the different approaches are?
Thanks a lot for your support!
Best Regards,
Daniel Hoppe
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
