LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (October 2011, week 4)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Tue, 25 Oct 2011 11:30:23 -0400
Reply-To:     "Jonas V. Bilenas" <jonas@JONASBILENAS.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Jonas V. Bilenas" <jonas@JONASBILENAS.COM>
Subject:      Re: SAS code to Bin Numeric Variables and Cross Tab across Target
              Variable
Content-Type: text/plain; charset=ISO-8859-1

You can check the code in a presentation I did at PhilaSUG and NJSUG:

http://jonasbilenas.com/PROC-RANK_PhilaSUG_Spring2009.pdf

Of course monotonicity is not a requirement for credit models or logistic models. You can transform the data (not WOE) with restricted cubic splines to model with. Note that you should never bin continuous data in the first place since you have a number of statistical issues doing that.

Splines, check another presentation: http://jonasbilenas.com/Loess.pdf

On Mon, 24 Oct 2011 20:53:03 -0700, .. <ajayohri@YAHOO.COM> wrote:

>Dear List

I am trying to do Logistic Regression (sighs..), I have hundreds of variables to narrow down. I am trying to do

1) Box Plots of each variable (just box plot not the usual and sometimes un �neccesary output in Proc Univariate) 2) Create Bins automatically of numeric variables so as to see if the target variable cross tab is increasing or decreasing monotonically.

I saw a link to Optibin Macro that is now in active.

Does anyone have any SAS code for automating the binning process?

Regards,

Ajay


Back to: Top of message | Previous page | Main SAS-L page