Date: Fri, 23 Dec 2005 14:31:52 -0500
Reply-To: "Nick ." <ni14@MAIL.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: "Nick ." <ni14@MAIL.COM>
Subject: PROC LOGIST MISSING VALUS get no scores---WHY
Content-Type: text/plain; charset="iso-8859-1"
I have a logistic model with 5 inputs. Some of these inputs are numeric and some are categories like A, B, C, etc. (20 such categories all the way up to U or V). If an observation has a mising value in only one of the 10 inputs that observation will not get scored. WHY? EXAMPLE: Suppose all 4 of the 5 inputs are 100% populated. Suppose further that 10% of the 5th variable is missing values here and there. Those observations will not get scored even if the remaining 4 inputs are populated. How can I get SAS to score them? I mean, if the boss gives me a dataset of 100K records to score and 10% of them are missing only a few values here and there, then SAS will only score 90K. The boss will then ask: What happended to the other10K?
Play 100s of games for FREE! http://games.mail.com/