[R] using the design matrix to correctly configure contrasts

[Karl Brand]

[Reposting to list]

Thank you Rich.

I still don't get it.

If i understand correctly- the MMC example on maize data outlines an 
issue with the multiple ties between group means. As a result of this, 
"user specified contrasts" needed specifying.

Its this part, understanding the design matrix (ie., the 'treatment 
contrast' for the ANOVA &/or lm design) so that one knows *how* to set 
up cunstom contrasts- the underlying logic- that i need to learn. Demystify.

In this maize example, if it is explained, regretfully i missed it. 
Perhaps i'm making somthing out of nothing, its self evident and i'm 
just complicating it. In that case a bucket of cold water is what i 
need. But maybe you have another example? Or an explicit tutorial 
covering this?

Unfortunatley the local stats here are thin on the ground and thus 
overworked. No time for "R-101" for troublsome biologists like me...

Back to the miaze example in case i missed 'it', and with sincere thanks 
for your suggestion,

karl



On 6/1/2010 6:03 PM, RICHARD M. HEIBERGER wrote:
 >
 > Please look at the maiz example, the last example in ?MMC in the HH 
package.
 > Follow the example all the way to the end.  It illustrates a problem and
 > then the resolution
 > of the problem.
 > install.packages("HH")  ## if you don't have it yet.
 > library(HH)
 > ?MMC
 > Rich

-- 
Karl Brand <k.brand at erasmusmc.nl>
Department of Genetics
Erasmus MC
Dr Molewaterplein 50
3015 GE Rotterdam
P +31 (0)10 704 3409 | F +31 (0)10 704 4743 | M +31 (0)642 777 268


On 6/1/2010 6:03 PM, RICHARD M. HEIBERGER wrote:
>
> Please look at the maiz example, the last example in ?MMC in the HH package.
> Follow the example all the way to the end.  It illustrates a problem and
> then the resolution
> of the problem.
> install.packages("HH")  ## if you don't have it yet.
> library(HH)
> ?MMC
> Rich

-- 
Karl Brand <k.brand at erasmusmc.nl>
Department of Genetics
Erasmus MC
Dr Molewaterplein 50
3015 GE Rotterdam
P +31 (0)10 704 3409 | F +31 (0)10 704 4743 | M +31 (0)642 777 268