To use PROC MIANALYZE to combine the covariance parameter estimates from PROC MIXED, sort the CovParms data set by the variable COVPARM before invoking MIANALYZE. If you use the SUBJECT= option in the RANDOM statement of PROC MIXED, use a DATA step to create a variable that concatenates the COVPARM and SUBJECT variables, then sort the data set by this new variable before invoking MIANALYZE.

This example comes from a pharmaceutical stability data simulation performed by Obenchain (1990). The observed responses are replicate assay results (in percent of label claim) at various shelf ages (in months).

      data rc; 
        input Batch Month @@; 
        do i = 1 to 6; 
          input Y @@; 
          output; 
        end; 
        datalines; 
         1   0  101.2 103.3 103.3 102.1 104.4 102.4 
         1   1   98.8  99.4  99.7  99.5    .     . 
         1   3   98.4  99.0  97.3  99.8    .     . 
         1   6  101.5 100.2 101.7 102.7    .     . 
         1   9   96.3  97.2  97.2  96.3    .     . 
         1  12   97.3  97.9  96.8  97.7  97.7  96.7 
         2   0  102.6 102.7 102.4 102.1 102.9 102.6 
         2   1   99.1  99.0  99.9 100.6    .     . 
         2   3  105.7 103.3 103.4 104.0    .     . 
         2   6  101.3 101.5 100.9 101.4    .     . 
         2   9   94.1  96.5  97.2 95.6     .     . 
         2  12   93.1  92.8  95.4 92.2   92.2  93.0 
         3   0  105.1 103.9 106.1 104.1 103.7 104.6 
         3   1  102.2 102.0 100.8  99.8    .     . 
         3   3  101.2 101.8 100.8 102.6    .     . 
         3   6  101.1 102.0 100.1 100.2    .     . 
         3   9  100.9  99.5 102.2 100.8    .     . 
         3  12   97.8  98.3  96.9  98.4  96.9  96.5 
         ;

PROC MI is invoked to perform mutiple imputation of Y using the monotone regression method.

      proc mi data=rc out=rc_impute seed=123;
        class batch;
        monotone regression(y=month batch);
        var batch month y;
        run;

Example 1: RANDOM statement without SUBJECT= option

In this example of PROC MIXED, the SUBJECT= option in the RANDOM statement is not used.

      proc mixed data=rc_impute covtest; 
        by _imputation_;
        model Y =; 
        random Month;
        ods output covparms=cvparms1;
        run;

The following step sorts the CovParms data set by the COVPARM variable.

      proc sort data=cvparms1;
         by covparm;
         run;

PROC MIANALYZE can then be invoked using the COVPARM variable in the BY statement.

      proc mianalyze data=cvparms1;
         by covparm;
         modeleffects estimate;
         stderr stderr;
         run;

Example 2: RANDOM statement using the SUBJECT= option

In this example, the model in PROC MIXED uses the SUBJECT= option in the RANDOM statement.

      proc mixed data=rc_impute covtest;
         by _imputation_;
         class Batch;
         model y = Month;
         random Int Month / type=un sub=Batch s; 
         ods output covparms=cvparms;
         run;

The following DATA step creates a new variable, COVPARM2, that concatenates the COVPARM and SUBJECT variables. The resulting data set is then sorted by the new variable.

      data cvparms;
         set cvparms;
         covparm2=covparm||subject;
         run;
      proc sort data=cvparms;
         by covparm2;
         run;

PROC MIANALYZE can then be invoked using the new variable in the BY statement.

      proc mianalyze data=cvparms;
         by covparm2;
         modeleffects estimate;
         stderr stderr;
         run;

Operating System and Release Information