[R] Output of Anova (CAR package) in Sweave

[John Fox]

Dear Maria,

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
On
> Behalf Of Maria Wolters
> Sent: June-16-09 6:31 AM
> To: r-help at r-project.org
> Subject: [R] Output of Anova (CAR package) in Sweave
> 
> Dear list,
> 
> I use Sweave almost exclusively for writing papers, and I have become
> quite spoiled by the excellent xtable export facilities. Has anybody
> written an xtable method for the Anova function in CAR, or has anybody
> used a different set of functions to import Anova results into
> a table in an Sweave document? If not, any handy hints on how to
> write a good homebrew based on the output of Anova would be much
> appreciated - in particular, where can I find Pillai test statistics
> and the formula for computing approximate F that is used in the
> CAR package? I have Fox' "R and S-Plus Companion to Regression".

I'm afraid that I can't help you with xtable(). 

To answer your other questions, Anova.mlm() was introduced relatively
recently, much after the 2002 publication of my R and S-PLUS Companion, and
so isn't described in the book. Approximate F-values for Pillai trace
statistics are computed using stats:::Pillai (i.e., the Pillai() function in
the stats package); you can simply examine the code for this function. 

I'm not entirely sure what you mean when you ask where you can find Pillai
test statistics. If you mean where in the code is Pillai's trace computed,
when you print an Anova.mlm object, the computation is done partly in
car:::print.Anova.mlm and partly in stats:::Pillai; when you summarize an
Anova.mlm.object, the computation is done in car:::summary.Anova.mlm,
car:::linearHypothesis.mlm, and stats:::Pillai. If you mean where can you
read about Pillai's trace test statistic, it's discussed in most texts that
cover multivariate linear models. Peter Dalgaard's article in R News
<http://cran.r-project.org/doc/Rnews/Rnews_2007-2.pdf> on R functions for
multivariate linear models provides some references.

I hope that this helps,
 John

> 
> I'm aware that this is a very basic question -
> thank you for your patience!
> 
> Maria Wolters
> 
> 
> 
> --
> Dr Maria Wolters
>     CSTR / HCRC, School of Informatics
>     MATCH Research Fellow
>     EdSST Project Manager
> mwolters AT staffmail.ed.ac.uk
> 
> --
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
> 
> ______________________________________________
> 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.