How do I get proc glimmix to recognize my numeric data with missing values?

[umrodger]

Hi all,

When I run the following code:

proc glimmix method=laplace data = work.April;

        class site plot YEAR WDENS;

        model BAIS = DLG /solution dist=poisson DDFM=BETWITHIN CL alpha=0.1;

        random intercept /subject=site TYPE=vc;

Run;

I get the error:

ERROR: Variable DLG should be either numeric or in CLASSES list.

The variable DLG is numeric, but there are missing values (".").  When I try the same code on the same variable with the missing values removed it works.  The problem is that the missing values are actually important, expecially when I have other models that are global.  Is there a way I can get this to work with the missing values still included?

Thanks!

[PGStats]

Could be related to a GLIMMIX bug :http://support.sas.com/kb/30/918.html

PG