[R] [FORGED] hier.part limitation

[Abdel-Rahman, Elfatih]

Dear Rolf Turner,

Thank you so much for your prompt and useful reply.

Regards
Elfatih

-----Original Message-----
From: Rolf Turner [mailto:r.turner at auckland.ac.nz] 
Sent: Monday, April 10, 2017 11:39 PM
To: Abdel-Rahman, Elfatih
Cc: r-help at stat.math.ethz.ch
Subject: Re: [FORGED] [R] hier.part limitation

On 10/04/17 21:33, Abdel-Rahman, Elfatih wrote:
> Dear all,
>
> Does the package "hier.part" limited to only 12 predictor variables.
> I  was trying to use it to partition (select) the most relevant 
> predictor variables our of 30 ones, and I always get an error message says:
> "Error: Number of variables must be < 13 for current implementation"

Uh, why do you suppose that the function puts out that error message?

If you want a function that handles 13 or more variables you'll have to write it yourself.

cheers,

Rolf Turner

P. S. I have no knowledge whatever of hier.part(), so I have no insight as to why that restriction is there, but I imagine there's a good reason, for some value of "good".

R. T.

--
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276