topic Re: Constraining the correlation of two random effects in proc MIXED in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/Constraining-the-correlation-of-two-random-effects-in-proc-MIXED/m-p/105012#M5550 <HTML><HEAD></HEAD><BODY><P>I don't know whether this will work, but it is possible to specify the covariance structure as being based on the correlations, using type=unr.&nbsp; From there, it seems possible that you could constrain a correlation to be a fixed value.</P><P></P><P>Steve Denham</P></BODY></HTML> Mon, 11 Feb 2013 13:16:26 GMT SteveDenham 2013-02-11T13:16:26Z Constraining the correlation of two random effects in proc MIXED https://communities.sas.com/t5/Statistical-Procedures/Constraining-the-correlation-of-two-random-effects-in-proc-MIXED/m-p/105011#M5549 <HTML><HEAD></HEAD><BODY><P>Hi all, </P><P>Does anybody know how I can constrain the correlation between two random effect to be 1?</P><P>I know about the PARMS statement (with the hold option), but with this statement I can constrain the Co-variance, but is there a direct way to constrain the correlation itself?</P><P></P><P></P><P>Thanks a lot </P><P>Eran.&nbsp; </P></BODY></HTML> Mon, 11 Feb 2013 13:00:47 GMT https://communities.sas.com/t5/Statistical-Procedures/Constraining-the-correlation-of-two-random-effects-in-proc-MIXED/m-p/105011#M5549 Eran 2013-02-11T13:00:47Z Re: Constraining the correlation of two random effects in proc MIXED https://communities.sas.com/t5/Statistical-Procedures/Constraining-the-correlation-of-two-random-effects-in-proc-MIXED/m-p/105012#M5550 <HTML><HEAD></HEAD><BODY><P>I don't know whether this will work, but it is possible to specify the covariance structure as being based on the correlations, using type=unr.&nbsp; From there, it seems possible that you could constrain a correlation to be a fixed value.</P><P></P><P>Steve Denham</P></BODY></HTML> Mon, 11 Feb 2013 13:16:26 GMT https://communities.sas.com/t5/Statistical-Procedures/Constraining-the-correlation-of-two-random-effects-in-proc-MIXED/m-p/105012#M5550 SteveDenham 2013-02-11T13:16:26Z