[R] Pseudo F statistics with index.G1

[Allan Engelhardt]

I guess that in R you have to be explicit about what you want to do.  
You can't just drop them, so you'll have to assign them some (other) 
value.  Try which(table(C)==1) to give you the values you need to change 
and then decide what to change them to.  The SAS documentation may tell 
you what it does; otherwise,  I'd suggest you look at the source, but I 
guess that is not an option for you.  Welcome to R.

Allan

On 06/07/10 15:48, Henrik Aldberg wrote:
> Thank you Allan,
>
> As i understand it the index.G1 function does not work if one of the 
> clusters in the partition only contains one object. Is there a way to 
> get around this in R? In SAS the PSF function seems to ignore the 
> presence of singleton clusters.
>
>
> Sincerely
>
> Henrik
>
>
>