[R] variance of a linear model

[Tim Hannig]

Hi,

this question may be off topic:

the unbiased estimator of the variance of the errors in a
linear regression moedel with p coefficients is:

	sigma2=sum((y-yi)^2)/(length(y)-p-1)

But what if i estimate transformations of the dependent an
independent variables (e.g. Box-Cox) too? May I calculate
the variance using

	sigma2=sum((y-yi)^2)/(length(y)-2*p-1)

or should I use the first formula then?

Thank you,
Tim
=================================================
Where Do You Want To Go Tomorrow?
http://www.GoPenguin.com
GoPenguin Network Inc.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._