Re: No splits using rpart?

[Volker Bahn <lochapoka@web.de>]

Look into the stopping criteria of the two methods. If I remember

correctly, their defaults are different. Otherwise the two methods give 
in my experience virtually identical results.
HTH

Volker

gerald.jean@dgag.ca wrote:
> Hello,
>
> I am trying to use rpart from the rpart library with little success so far,
> I have used it in the past though for similar problems but I can't figure
> out what I am doing wrong this time, no matter what I do, how I simplify
> the problem, I don't get any splits.  Here is a simplified run with number
> of rows considerably reduced (nsinB2 is a binary variable, 0 or 1):
>
>   
> > test.rpart <- rpart(formula =s.factor(nsinB2) ~ priv + kmprom + ageveh,
> >     
> +                               data =est2)
>   
> > test.rpart
> >     
> n=05895
>
> node), split, n, loss, yval, (yprob)
>       * denotes terminal node
>
> 1) root 105895 32701 0 (0.6911941 0.3088059) *
>
>
> The builtin tree function has no problem with the same call, it gives me
> lots of splits:
>
>   
> > test.tree <- tree(formula =s.factor(nsinB2) ~ priv + kmprom + ageveh,
> >     
> +                   data =est2)
>   
> > test.tree
> >     
> node), split, n, deviance, yval, (yprob)
>       * denotes terminal node
>
>      1) root 105895 1.309e+05 0 ( 0.6912 0.30880 )
>        2) ageveh<6.5 77648 9.865e+04 0 ( 0.6685 0.33150 )
>          4) ageveh<3.5 47696 6.150e+04 0 ( 0.6543 0.34570 )
>            8) ageveh<-0.5 3331 3.901e+03 0 ( 0.7277 0.27230 )
>             16) priv: Non 2856 3.408e+03 0 ( 0.7160 0.28400 )
>               32) ageveh<-1.5 108 9.732e+01 0 ( 0.8333 0.16670 ) *
>               33) ageveh>-1.5 2748 3.302e+03 0 ( 0.7114 0.28860 )
>                 66) kmprom<17250 1736 2.055e+03 0 ( 0.7212 0.27880 )
> .
> .
> .
>
> and the output continues for lots more splits.  Anyone can figure out what
> I am doing wrong?
>
> Thanks for your support,
>
> Gérald Jean
> Conseiller senior en statistiques, Actuariat
> télephone            : (418) 835-4900 poste (7639)
> télecopieur          : (418) 835-6657
> courrier électronique: gerald.jean@dgag.ca
>
> "In God we trust, all others must bring data"  W. Edwards Deming
>
>
>
> Le message ci-dessus, ainsi que les documents l'accompagnant, sont destinés
> uniquement aux personnes identifiées et peuvent contenir des informations
> privilégiées, confidentielles ou ne pouvant être divulguées. Si vous avez
> reçu ce message par erreur, veuillez le détruire.
>
> This communication (and/or the attachments) is intended for named
> recipients only and may contain privileged or confidential information
> which is not to be disclosed. If you received this communication by mistake
> please destroy all copies.
>
>
>   

--

--------------------------------------
Volker Bahn
Department of Biology
McGill University
Stewart Biol. Bldg. W3/5
1205 ave Docteur Penfield
Montreal, QC, H3A 1B1
Canada
t: (514) 398-6428
f: (514) 398-5069
volker.bahn@mcgill.ca
www.volkerbahn.com

Lat-Long:       
45.50285, -73.5814
--------------------------------------