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 (May 2007, week 1)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Wed, 2 May 2007 08:59:56 -0400
Reply-To:     Raghu Venkat <raghustays@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Raghu Venkat <raghustays@GMAIL.COM>
Subject:      Re: Which SAS procedure to use
Comments: To: kunalpkelkar@aim.com
In-Reply-To:  <8C95AD01040B4E4-16D0-39BC@webmail-db06.sysops.aol.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

try to first see whats in the data. a correlation matrix using proc corr might be a good starting point to see if any multicollinearity issues exist. Then see if you can combine some of those variables and reduce it some other variable that might make sense.

i dont think anyone here will have a ready made solution to your question. This sounds like one of the class projects :)

- Raghu

On 5/2/07, Kunal Kelkar <kunalpkelkar@aim.com> wrote: > > Hi All, > > I am having a SAS dataset with close to 640 > variables out of which 430 are of categorical type.I am trying to fit > regression model using this data. My response variable is also of > categorical type.Many variables are having lot of missing values.I need > help to find out how I can reduce the number of variables and how can tackle > the null values.Also which SAS procedure will be helpful for this. If > possible if anyone can provide some example that will be great. > > Thanks in advance. > > Thanks, > Kunal > > ________________________________________________________________________ > Check Out the new free AIM(R) Mail -- 2 GB of storage and industry-leading > spam and email virus protection. >


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