|Date: ||Tue, 7 Jun 2005 09:39:16 -0400|
|Reply-To: ||Kevin Roland Viel <kviel@EMORY.EDU>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||Kevin Roland Viel <kviel@EMORY.EDU>|
|Subject: ||residuals: correlation with DV|
|Content-Type: ||TEXT/PLAIN; charset=US-ASCII|
I have performed an analysis using a non-SAS program on the members of
21 pedigrees. My outcome variable was FVIII.
My plot of the residuals (FVIII - FVIII_predicted) versus FVIII appears
to suggest a correlation, i.e. subjects with a high FVIII have larger
residuals and those with low FVIII have smaller residuals. This concerned
me. Is my concern warranted? The residual plots versus two independent
variables appears random, as I expected.
Department of Epidemiology
Rollins School of Public Health
Atlanta, GA 30322