[R] binning by frequency

[William Dunlap]

You might look at the 'mdlp' function in the 'discretization' package.
(SPSS has a procedure called 'optimal binning' that uses the 'minimum
description length principle' to do the binning.)

Bill Dunlap
Spotfire, TIBCO Software
wdunlap tibco.com


> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf
> Of Santosh
> Sent: Tuesday, November 27, 2012 11:59 AM
> To: r-help
> Subject: Re: [R] binning by frequency
> 
> Thanks for your response.
> Was wondering if there are any R functions/packages to perform optimal
> binning of continuous data.
> 
> Thanks, again.
> Santosh.
> 
> On Tue, Nov 27, 2012 at 9:09 AM, Mark Lamias <mlamias at yahoo.com> wrote:
> 
> > You might find the binning function in the sm package helpful here.
> >
> > --Mark Lamias
> >
> >   ------------------------------
> > *From:* Santosh <santosh2005 at gmail.com>
> > *To:* r-help <r-help at r-project.org>
> > *Sent:* Tuesday, November 27, 2012 9:59 AM
> > *Subject:* [R] binning by frequency
> >
> > Dear Rxperts,
> >
> > is there way to identify intervals from continuous data (having some kind
> > of a pattern) and then pick the value of most frequency?
> >
> >
> > a1 <- round(rnorm(50,mean=0,0.1),2)
> > a2 <- round(rnorm(50,mean=1,0.2),1)
> > a3 <- round(rnorm(50,mean=5,1),0)
> > a4 <- round(rnorm(50,mean=14,4),0)
> > a5 <- round(rnorm(50,mean=30,8),0)
> >
> > b1 <- rbind(a1,a2,a3,a4,a5)
> >
> > hist(b1,brea=100) # shows intervals and values with varying frequency.
> >
> > unlike the mean values of a1.... a5 above, I don't know the nominal values.
> > I would like an algorithm to identify intervals and pick the value with
> > most frequency.
> >
> > I tried "cut", "split" and was not successful.
> > Any suggestions/tips are highly welcome.
> > Thanks and regards,
> > Santosh
> >
> >     [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> > http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> >
> >
> >
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.