Hi all,
I obtained an error message "Within-imputation Estimate missing for effect PGS_MEGALGC_APP_APOE*TIME in _Imputation_= 1 in the inputPARMS= data set" when I run the following code:
PROC MIXED DATA = IMP_IMM_LONG METHOD = RMEL; CLASS DBID FAMILY TIME(REF="2") GENDER; MODEL CS_IMM_LRN = PGS_MEGALGC_APP_APOE PGS_MEGALGC_APP_APOE*TIME AGEATVISIT TIME GENDER EDYEARS_CODED_MAX20 PC1 PC2 PC3 PC4 PC5 / CHISQ S DDFM = BW; RANDOM INTERCEPT / TYPE = UN SUBJECT = FAMILY G; REPEATED TIME / TYPE = CS SUBJECT = DBID(FAMILY) R; BY _IMPUTATION_; ODS OUTPUT SOLUTIONF = APP_IMM_BETA; RUN; PROC MIANALYZE PARMS(CLASSVAR=FULL) = APP_IMM_BETA; CLASS TIME GENDER; MODELEFFECTS PGS_MEGALGC_APP_APOE PGS_MEGALGC_APP_APOE*TIME AGEATVISIT TIME GENDER EDYEARS_CODED_MAX20 PC1 PC2 PC3 PC4 PC5; RUN;
This error appears after SAS executed the PROC MIANALYZE procedure.
The parms data set I used for the PROC MIANALIZE procedure looks like this:
What should I do to avoid this error message? Any help is appreaciated!
I realized the problem was my variable name is too long. After I change the name to a shorter one, the error message disappear.
I realized the problem was my variable name is too long. After I change the name to a shorter one, the error message disappear.
SAS Innovate 2025 is scheduled for May 6-9 in Orlando, FL. Sign up to be first to learn about the agenda and registration!
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.