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 1998, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Mon, 19 Oct 1998 12:10:09 -0500
Reply-To:     Cherilyn Young <cyoung@UTS.CC.UTEXAS.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Cherilyn Young <cyoung@UTS.CC.UTEXAS.EDU>
Subject:      Classification lines in SAS?
Content-Type: TEXT/PLAIN; charset=US-ASCII

Hi everyone,

Is there a SAS function for drawing ideal classification lines (dividing lines) between 2 groups of variables on a scatterplot?

I'm dealing with a scatterplot situation (simple 2-var scatterplot), where each subject (data point) is classified by one value each of 2 dichotomous variables. I want to find where the 2 classification lines are, which would divide the scatterplot into 4 regions.

I've done the variables' coding and the plot on Minitab already as a "late-night inspiration" estimate, but I need to have the classification boundary lines drawn in and figured out (like regression lines) instead of just estimating them visually. I'm not necessarily wanting the classification boundary lines to go according to my original coding of the variables, but how the stats work out (any differences between my coding and statistical (mis-)categorization are welcome).

If SAS can't do it, does anyone know of any other software (psychometric, epidemiological, etc.) that can do this?

Thank you in advance,

Cherilyn Young


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