[R] Verify the linear regression model used in R ( fundamental theory)

[Gabor Grothendieck]

On Tue, Jun 22, 2010 at 7:46 PM, Yi <liuyi.feier at gmail.com> wrote:
> Hi, folks,
>
> As I understand, Least-squares Estimate (second-moment assumption) and the
> Method of Maximum Likelihood (full distribtuion assumption) are used for
> linear regression.
>
> I do >?lm, but the help file does not tell me the model employed in R. But
> in the book 'Introductory Statistics with R', it indicates R estimate the
> parameters using the method of Least-squares. However it assumes the error
> is iid N(o,sigma^2).
>
> Am I correct? Is there any general way (like RSiteSearch() ) to find out
> what the model (theory) is for specific function? Let's say how to find out
> the assumption and the model used for rlm.
>
> Thanks
>
> Yi

If you look on the ?lm page under See Also it says that the underlying
low level function is lm.fit.   On the ?lm.fit page it says that lm
uses least squares.

Systematic following of See Also will often get you the info you need
if its not already on the help page.

In the  case of packages there may also be a vignette and/or a package
home page with the info.  library(help = "mypackage") will typically
list mypackage's home page, if any, on the URL: line  and the
vignettes, if any, will be listed at the end.

The CRAN web page for the mypackage package can be found by googling
for:   CRAN mypackage.