Error: All replicates have absent response levels or non-convergence status. In PROC SURVEYLOGISTIC

[PMA1]

Hi, I am analyzing the California Health Interview Survey (CHIS) 2014-2015 combined data and would like to run

PROC SURVEYLOGISTICS (take complex design into consideration) in SAS with an interaction term using LSMEANS statement.

I got this error message when running the following codes:

"ERROR: All replicates have absent response levels or non-convergence status."

Does anyone know what does the error message means, and how to fit this? Thanks!

PROC SURVEYLOGISTIC DATA = work.combined_recoded VARMETHOD=JACKKNIFE;

WEIGHT fnwgt0;

REPWEIGHT fnwgt1-fnwgt160/JKCOEFS=1;

CLASS R_SMKCUR (REF="NOT CURRENT SMOKER")

R_RaceEthnicities (REF="Whites")

R_DSTRS30 (REF="NO")

SRAGE_P1 (REF="18-25 YEARS")

AHEDC_P1 (REF="GRADE 12/H.S. DIPLOMA")

R_Sex (REF="FEMALE")/PARAM=REF;

MODEL R_SMKCUR = R_DSTRS30 R_RaceEthnicities R_Sex SRAGE_P1 AHEDC_P1 R_RaceEthnicities*R_DSTRS30/ LINK=glogit;

lsmeans R_RaceEthnicities*R_DSTRS30 /oddsratio cl diff;

RUN;

[SAS_Rob]

I believe that message is essentially saying that none of the replicate models converged.  I suspect you probably have some sort of separation in your model, leading to non-covergence.  Most likely the interaction is perfectly predicting the response--you could try running a Proc FREQ with 

tables R_SMKCUR*R_RaceEthnicities*R_DSTRS30;

to diagnose it, or remove the interaction from the model and see if the problem goes away.