https://communities.sas.com/t5/Statistical-Procedures/Collinearity-diagnostics/m-p/291632#M15516 <P><A href="http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_reg_details24.htm" target="_blank">Proc REG documentation</A> states:</P> <P>&nbsp;</P> <PRE>A collinearity problem occurs when a component associated with a high condition index contributes strongly (variance proportion greater than about 0.5) to the variance of two or more variables. </PRE> <P>each line in the collinearity diagnostics table is about a component. Condition indexes inferior to 10 are not high.</P> <P>&nbsp;</P> <P>It also says:</P> <P>&nbsp;</P> <PRE>There are no formal criteria for deciding if a VIF is large enough to affect the predicted values. </PRE> Mon, 15 Aug 2016 03:39:11 GMT PGStats 2016-08-15T03:39:11Z https://communities.sas.com/t5/Statistical-Procedures/Collinearity-diagnostics/m-p/291605#M15508 <P>Need a refresher/reminder about collinearity diagnostics.&nbsp;</P><P>&nbsp;</P><P>I know collinearity is present if there is a high condition index and VIFs that are greater than 0.5 of two or more variables.</P><P>&nbsp;</P><P>When it discusses "two or more variables" is that on the same line or is that for all of the VIFs?</P><P>&nbsp;</P><P>I have attached by output - just can't remember if it's VIFs in the same line or all around in the table.</P><P>&nbsp;</P><P>&nbsp;</P> Sun, 14 Aug 2016 18:53:21 GMT https://communities.sas.com/t5/Statistical-Procedures/Collinearity-diagnostics/m-p/291605#M15508 buder 2016-08-14T18:53:21Z https://communities.sas.com/t5/Statistical-Procedures/Collinearity-diagnostics/m-p/291632#M15516 <P><A href="http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_reg_details24.htm" target="_blank">Proc REG documentation</A> states:</P> <P>&nbsp;</P> <PRE>A collinearity problem occurs when a component associated with a high condition index contributes strongly (variance proportion greater than about 0.5) to the variance of two or more variables. </PRE> <P>each line in the collinearity diagnostics table is about a component. Condition indexes inferior to 10 are not high.</P> <P>&nbsp;</P> <P>It also says:</P> <P>&nbsp;</P> <PRE>There are no formal criteria for deciding if a VIF is large enough to affect the predicted values. </PRE> Mon, 15 Aug 2016 03:39:11 GMT https://communities.sas.com/t5/Statistical-Procedures/Collinearity-diagnostics/m-p/291632#M15516 PGStats 2016-08-15T03:39:11Z