[R] Minimum valid number of observations for rpart

[Prof Brian Ripley]

On Fri, 30 Mar 2007, fjlozl at unileon.es wrote:

> Hi,
> I wonder if anyone knows a study dealing with the minimum valid number of
> observations when using CART?.

I have no idea what you mean by 'valid' here.
Could you answer the question for logistic regression to indicate to us 
what form of answer you are expecting?

> On top of that, when using RandomForest, is it possible to obtained a
> interpretable tree model as the graphical output of the analysis, just
> like in "rpart"?

By definition randomForest (sic) does not produce a tree but a forest.

> Thanks a lot in advance
>
> Javier Lozano
> Universidad de León
> Spain


-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595