Comparison proc glm and proc mixed

[Clg]

Dear all,

I would like to compare PROC GLM and PROC MIXED for one model and I was wondering if it was possible to get the correspondance of "covtest" in proc mixed with the prog glm?

My model with proc mixed is:

proc mixed data=ADPP covtest cl method=type3 alpha=0.1;
by PARAMN PARAMCD;
class TRTAN SEX USUBJID;
model AVAL = TRTAN SEX / solution ddfm=kr;
random USUBJID(SEX) / type=vc;
run;

My model with prog glm is:

proc glm data=ADPP ;
class TRTAN SEX USUBJID;
model AVAL = TRTAN SEX USUBJID(SEX) ;
random USUBJID(SEX) / test ;
run;

I obtain the same Mean Squares with the two models but with the proc GLM I don't have the covariance parameters with the interval of confidence.

Is-it possible to obtain these parameters with the proc glm ? If yes, how I can get them ?

Thank you in advance.

Best regards;

Clg

[SteveDenham]

From the PROC GLM documentation:

Note: PROC GLM uses only the information pertaining to expected mean squares when you specify the TEST option in the RANDOM statement and, even then, only in the extra F tests produced by the RANDOM statement. Other features in the GLM procedure—including the results of the LSMEANS and ESTIMATE statements—assume that all effects are fixed, so that all tests and estimability checks for these statements are based on a fixed-effects model, even when you use a RANDOM statement. Therefore, you should use the MIXED procedure to compute tests involving these features that take the random effects into account; see the section PROC GLM versus PROC MIXED for Random-Effects Analysis and Chapter 81: The MIXED Procedure, for more information.

This applies to interval estimates as well, so that GLM doesn't produce CI's as you might think of them.  I thought you might be able to add the CLPARMS and SOLUTION options to the MODEL statement and get something, but that appears to give CI's for each level of the effects.

I guess this all boils down to "Don't use PROC GLM for this particular task.  Use PROC MIXED."

SteveDenham

[Clg]

Thnak you for your answer.

[Ksharp]

I think it is not possible. As Steve said, PROC GLM is for fixed effect model,PROC MIXED is for mixed effect model . Since COVTEST is a test for mixed effect (is signifcant or not), there is no reason to use PROC GLM .