[R] p-value for non linear model

[Bert Gunter]

Tatiana:

It sounds like you are in way over your head statistically. This is
not a statistics tutorial site (though sometimes good folks do help
witjh this). I suggest you try http://stats.stackexchange.com/  .
Better yet, work with your local statistician.

Cheers,
Bert

On Thu, Sep 15, 2011 at 6:09 AM, Tatiana Donnay
<tatiana.donnay at gmail.com> wrote:
> Hello,
>
> I want to understand how to tell if a model is significant.
>
> For example I have vectX1 and vectY1.
> I seek first what model is best suited for my vectors and
> then I want to know if my result is significant.
>
> I'am doing like this:
>
> model1 <- lm(vectY1 ~ vectX1, data= d),
>
> model2 <- nls(vectY1 ~ a*(1-exp(-vectX1/b)) + c, data= d,
> start = list(a=1, b=3, c=0))
>
> aic1 <- AIC(model1)
> aic2 <- AIC(model2)
>
> if (aic1 < aic2) print("Model1 is better")
> else print("Model2 is better")
>
> for example aic1 < aic2 I'am doing summary(model1) and I have p-value to
> know if my result is significant.
>
> but if  aic2 < aic1, so model 2, non linear is better. I'am doing
> summary(model2) bur there isn't p-value. I read that it is normal. So how
> can I know if my resultat is significant?
>
> I can't use summary.nls(), because, I'm working with mac and it's no
> possible to install it.
>
>
> Thank you
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> 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.
>



-- 
"Men by nature long to get on to the ultimate truths, and will often
be impatient with elementary studies or fight shy of them. If it were
possible to reach the ultimate truths without the elementary studies
usually prefixed to them, these would not be preparatory studies but
superfluous diversions."

-- Maimonides (1135-1204)

Bert Gunter
Genentech Nonclinical Biostatistics
467-7374