R help - Oct 2005 - selecting every nth item in the data

I want to make a glm and then use predict.  I have a fairly small sample
(4000 cases) and I want to train on 90% and test on 10% but I want to do
it in slices so I test on every 10th case and train on the others.  Is
there some simple way to get these elements?
 
Stephen

-- 



21/10/2005
 

	[[alternative HTML version deleted]]

On 25 Oct 2005 at 13:48, Stephen Choularton wrote:

From:           	"Stephen Choularton" <mail at bymouth.com>
To:             	"R Help" <r-help at stat.math.ethz.ch>
Date sent:      	Tue, 25 Oct 2005 13:48:42 +1000
Subject:        	[R] selecting every nth item in the data
> I want to make a glm and then use predict.  I have a fairly small
> sample (4000 cases) and I want to train on 90% and test on 10% but I
> want to do it in slices so I test on every 10th case and train on the
> others.  Is there some simple way to get these elements?
As glm accept subet you probably can use

test<-seq(1,4000,n)

glm(...., subset = test)

to select only n'th item or

glm(...., subset = -test)

to select all but n'th items.

or you can split the data freme by

mydf[test,] or mydf[-test,]

HTH
Petr


> 
> Stephen
> 
> -- 
> 
> 
> 
> 21/10/2005
> 
> 
>  [[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
Petr Pikal
petr.pikal at precheza.cz