Re: [R] curve fit

From: Gabor Grothendieck <ggrothendieck_at_gmail.com>
Date: Fri 21 Oct 2005 - 15:17:54 EST

y ~ 1/(a + b*x^c) looks pretty good to me.

If we assume a starting value of c=1 then we can get the starting values of a and b from lm(1/y ~ x) and then do an nls fit.

On 10/20/05, Xiaodong Jin <close2ceo@yahoo.com> wrote:
> How to obtain the FUNCTION for the following smooth curve?
> x 0 100 250 500 1000 4000
> y 1.8 1.2 1.02 0.99 0.97 0.85
>
> Thanks,
> SJ
>
>
> ---------------------------------
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> 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
>



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 Received on Fri Oct 21 15:24:33 2005

This archive was generated by hypermail 2.1.8 : Sun 23 Oct 2005 - 19:15:35 EST