[R] covariance matrix of the regression coefficients

Doran, Harold HDoran at air.org
Mon Oct 29 22:37:02 CET 2007


I don't think you ever want to do it this way. vcov() gives what is
needed. But, if you perform matrix operations represented algebraically
as t(x) %*% x, then use crossprod(x) and NOT t(x) %*% x

See the paper at the link below for reasons why

@ARTICLE{Rnews:Bates:2004,
  author = {Douglas Bates},
  title = {Least Squares Calculations in {R}},
  journal = {R News},
  year = 2004,
  volume = 4,
  number = 1,
  pages = {17--20},
  month = {June},
  url = {http://CRAN.R-project.org/doc/Rnews/},
  pdf = {http://CRAN.R-project.org/doc/Rnews/Rnews_2004-1.pdf}
}

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of A. Beaujean
> Sent: Monday, October 29, 2007 4:25 PM
> To: Peter B. Mandeville
> Cc: r-help
> Subject: Re: [R] covariance matrix of the regression coefficients
> 
> If X is your p-1 variable matrix (with the first column 
> vector being 1s), i.e., nrow(X)=n and ncol(X)=p
> 
> then
> MSE<-summary(lm(Y~X[2]+X[3] + ...X[P-1]))$s^2
> 
> and your coefficient (co)variance matrix is
> MSE*ginv(t(X)%*%X)
> 
> Best,
> 
> Alex
> 
> 
> On 10/29/07, Peter B. Mandeville <mandevip at hotmail.com> wrote:
> >
> > Greetings,
> >
> > On page 273, Cohen, Cohen, West, and Aiken (Applied Multiple 
> > Regression/Correlation Analysis for the Behavioral Sciences, Third 
> > Edition", state that the covariance matrix of the regression 
> > coefficients is provided by standard programs for multiple 
> regression, 
> > including SAS, SPSS, and SYSTAT. How does one calculate the 
> matrix in R. Thank you very much.
> >
> > pbmPeter B. Mandeville cel:      444 860 3204 tel:  52 444 
> 826 2346-49 ext
> > 532 fax: 52 444 826 2352 P.D. Favor de confirmar la llegada de este 
> > correo. Gracias.
> > _________________________________________________________________
> >
> >
> >         [[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.
> >
> 
> 
> 
> --
> ***************
> A. Alexander Beaujean, Ph.D., LSSP
> Licensed Psychologist (Provisional, TX)
> http://myprofile.cos.com/abeaujean
> http://www.baylor.edu/soe/faculty/index.php?id=38476
> 
> 
> "General impressions are never to be trusted. Unfortunately 
> when they are of
> long standing they become fixed rules of life, and assume a 
> prescriptive
> right not to be questioned. Consequently those who are not 
> accustomed to
> original inquiry entertain a hatred and a horror of 
> statistics. They cannot
> endure the idea of submitting their sacred  impressions to 
> cold-blooded
> verification. But it is the triumph of scientific men to rise 
> superior to
> such superstitions, to devise tests by which the value of 
> beliefs may be
> ascertained, and to feel sufficiently masters of themselves to discard
> contemptuously whatever may be found untrue." --Sir Francis 
> Galton, FRS
> 
> 	[[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.
> 



More information about the R-help mailing list