R help - Oct 2006 - [Q] How to fit data to a straight line

Dear R users

I have a question about how to fit data to a straight line.
I tried nls to do it, but it didn't work.

The reason I want to fit data to a straight line is that I need to compare
AIC or BIC values of the two models (a straight line model vs a nonlinear
curve model). Fitting data to a nonlinear curve is straightforward, but I
could not figure out how to fit the data to a straight line and therefore, I
could not get the AIC/BIC values of this simple model. If I can get the
AIC/BIC value of the straight line fit without actually doing a fit, that is
also fine.

Thanks in advance.

Young-Jin

	[[alternative HTML version deleted]]

Young-Jin Lee <youngjin.michael <at> gmail.com> writes:
> 
> Dear R users
> 
> I have a question about how to fit data to a straight line.
> I tried nls to do it, but it didn't work.
> 
>
  You really need to read a statistics book!
R assumes that you have basic statistical training
and/or access to basic statistics texts (e.g.
books by Dalgaard or any of the other fine books
listed on the R web site, but *any* introductory
statistics textbook would do).  

  I also can't see why nls() wouldn't work.
> x = 1:10
> y = rnorm(10,1+x,sd=0.4)
> AIC(nls(y~a+b*x,start=list(a=1,b=1)))
[1] 11.3995
> AIC(lm(y~x))
[1] 11.3995

  I probably shouldn't have answered this --
please read the posting guide before you post
again ...

  sincerely,
  Ben Bolker