R help - Oct 2009 - Substituting the extracted coefficients into the formula, exctracted from the result of nls()

Dear all,

Here I come with another stupid question.  Suppose I want to use nls()
to fit a series of data (here modelled by generated points), then plot
the points and the fitting curve. I figured out some way of doing it:

x <- runif(1:20, 0, 10)
y <- 0.1*x^2 - rep(3, length(x)) + rnorm(length(x), sd = 0.5)

yfit <- nls(y ~ a*x^2 + b*x + c,
            start = list(a = 1, b = 1, c = 1),
            trace = TRUE)

plot(x, y)
curve(coef(yfit)[1]*x^2 + coef(yfit)[2]*x + coef(yfit)[3], 0, 10,
      add = TRUE)

However, this code is rather fortranesque, and most likely there
exists a much more elegant way in R/S, something like abline() which
exists for straight lines. Since both the formula and the coefficients
are available as a result of nls() (here as formula(yfit) and
coef(yfit)), I thought there ought to be a way something along the
following lines:

f <- substitute(formula(yfit), as.list(coef(yfit)))
curve(f, ...)

However, this snippet of code doesn't work. Am I thinking into the
correct direction at all?

Thanks in advance.

All the best,
Primo?

-- 
Primo? Peterlin,   In?titut za biofiziko, Med. fakulteta, Univerza v Ljubljani
Lipi?eva 2, SI-1000 Ljubljana, Slovenija.        primoz.peterlin at mf.uni-lj.si
Tel +386-1-5437612, fax +386-1-4315127,  http://biofiz.mf.uni-lj.si/~peterlin/
F8021D69 OpenPGP fingerprint: CB 6F F1 EE D9 67 E0 2F  0B 59 AF 0D 79 56 19 0F

Try with predict:

plot(x, y)
lines(0:10, predict(yfit, list(x = 0:10)))

2009/10/9 Primoz PETERLIN <primozz.peterlin at
gmail.com>:
> Dear all,
>
> Here I come with another stupid question. ?Suppose I want to use nls()
> to fit a series of data (here modelled by generated points), then plot
> the points and the fitting curve. I figured out some way of doing it:
>
> x <- runif(1:20, 0, 10)
> y <- 0.1*x^2 - rep(3, length(x)) + rnorm(length(x), sd = 0.5)
>
> yfit <- nls(y ~ a*x^2 + b*x + c,
> ? ? ? ? ? ?start = list(a = 1, b = 1, c = 1),
> ? ? ? ? ? ?trace = TRUE)
>
> plot(x, y)
> curve(coef(yfit)[1]*x^2 + coef(yfit)[2]*x + coef(yfit)[3], 0, 10,
> ? ? ?add = TRUE)
>
> However, this code is rather fortranesque, and most likely there
> exists a much more elegant way in R/S, something like abline() which
> exists for straight lines. Since both the formula and the coefficients
> are available as a result of nls() (here as formula(yfit) and
> coef(yfit)), I thought there ought to be a way something along the
> following lines:
>
> f <- substitute(formula(yfit), as.list(coef(yfit)))
> curve(f, ...)
>
> However, this snippet of code doesn't work. Am I thinking into the
> correct direction at all?
>
> Thanks in advance.
>
> All the best,
> Primo?
>
> --
> Primo? Peterlin, ? In?titut za biofiziko, Med. fakulteta, Univerza v
Ljubljani
> Lipi?eva 2, SI-1000 Ljubljana, Slovenija. ? ? ? ?primoz.peterlin at
mf.uni-lj.si
> Tel +386-1-5437612, fax +386-1-4315127,
?http://biofiz.mf.uni-lj.si/~peterlin/
> F8021D69 OpenPGP fingerprint: CB 6F F1 EE D9 67 E0 2F ?0B 59 AF 0D 79 56 19
0F
>
> ______________________________________________
> R-help at r-project.org 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.
>


-- 
Henrique Dallazuanna
Curitiba-Paran?-Brasil
25? 25' 40" S 49? 16' 22" O