Date:         Wed, 10 Nov 1999 11:59:56 +0100
Reply-To:     Torben Haslund <Torben.Haslund@VBIOL.SLU.SE>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Torben Haslund <Torben.Haslund@VBIOL.SLU.SE>
Subject:      Degrees of freedom in PROC MIXED
Content-Type: text/plain; charset="us-ascii"

Dear SAS-Lers

I want to use PROC MIXED (SAS 6.12) to compare different models based on
the same split plot design. What changes from one model to another is/are
the covariate(s) and their interactions with class factors in the MODEL
statement. The RANDOM statement is constant: RANDOM BLOCK(TRT);

Therefore I need to calculate the mean square error (MSE) and the total
mean square (MST) for further calculations of pseudo-R2, pseudo-R2adj,
pseudo-Q2 and pseudo-Q2adj:

R2         =1-   SSE/SST;
R2adj      =1-   MSE/MST;
Q2         =1-   SSE/SST    with jack-knifed predictions and residuals;
Q2adj      =1-   MSE/MST    with jack-knifed predictions and residuals;

Could any of you could give some advise on how to get appropriate
expressions from PROC MIXED for these DFs? - I would like to let a macro
extract the information from data sets made by the MAKE statement. It
should account for degrees of freedom for effects mentioned in the MODEL
and in the RANDOM statements.

 yours

torben haslund


:::::::::::::::::::::::::::::::::
Torben Haslund
Department of Plant Biology
Swedish University of Agricultural Sciences
P.O. Box 7080
S - 750 07 UPPSALA, Sweden

E-mail address:                 Torben.Haslund@vbiol.slu.se
Home and postal address:        Ostre Paradisvej 7B,
                                DK-2840 HOLTE, Denmark
Phone home:                     +45  4541 1198