ROC plot

The GENMOD Procedure

       Model Information

Data Set              WORK.SCHIZ
Distribution              Normal
Link Function           Identity
Dependent Variable             y


Number of Observations Read        1652
Number of Observations Used        1500
Missing Values                      152


                        Class Level Information

Class      Levels    Values

id            413    1103 1104 1105 1106 1107 1108 1109 1110 1111 1113
                     1114 1115 1118 1124 1129 1136 1140 1301 1302 1303
                     1304 1305 1306 1307 1308 1309 1311 1312 1313 1314
                     1315 1316 1317 1318 1321 1329 2103 2104 2105 2106
                     2107 2108 2109 2110 2111 2112 2113 2115 2116 2117
                     ...


    Parameter Information

Parameter       Effect

Prm1            Intercept
Prm2            drug
Prm3            sqrtweek
Prm4            drug*sqrtweek


           Criteria For Assessing Goodness Of Fit

Criterion                 DF           Value        Value/DF

Deviance                1496       2221.3907          1.4849
Scaled Deviance         1496       1500.0000          1.0027
Pearson Chi-Square      1496       2221.3907          1.4849
Scaled Pearson X2       1496       1500.0000          1.0027
Log Likelihood                    -2422.9091


Algorithm converged.
ROC plot

The GENMOD Procedure

                Analysis Of Initial Parameter Estimates

                                Standard   Wald 95% Confidence      Chi-
Parameter       DF   Estimate      Error          Limits          Square

Intercept        1     5.3961     0.1070     5.1864     5.6058   2543.24
drug             1    -0.0098     0.1229    -0.2508     0.2311      0.01
sqrtweek         1    -0.4048     0.0738    -0.5494    -0.2602     30.10
drug*sqrtweek    1    -0.5243     0.0839    -0.6887    -0.3599     39.05
Scale            1     1.2169     0.0222     1.1742     1.2613

   Analysis Of Initial
   Parameter Estimates

Parameter       Pr > ChiSq

Intercept           <.0001
drug                0.9363
sqrtweek            <.0001
drug*sqrtweek       <.0001
Scale

NOTE: The scale parameter was estimated by maximum likelihood.


             GEE Model Information

Correlation Structure               Independent
Subject Effect                  id (413 levels)
Number of Clusters                          413
Clusters With Missing Values                101
Correlation Matrix Dimension                  4
Maximum Cluster Size                          4
Minimum Cluster Size                          1


Algorithm converged.


               Analysis Of GEE Parameter Estimates
                Empirical Standard Error Estimates

                       Standard   95% Confidence
Parameter     Estimate    Error       Limits            Z Pr > |Z|

Intercept       5.3961   0.0882   5.2233   5.5689   61.19   <.0001
drug           -0.0098   0.1024  -0.2106   0.1909   -0.10   0.9236
sqrtweek       -0.4048   0.0657  -0.5335  -0.2761   -6.16   <.0001
drug*sqrtweek  -0.5243   0.0763  -0.6738  -0.3748   -6.87   <.0001
ROC plot

The GENMOD Procedure

               Analysis Of GEE Parameter Estimates
               Model-Based Standard Error Estimates

                       Standard   95% Confidence
Parameter     Estimate    Error       Limits            Z Pr > |Z|

Intercept       5.3961   0.1071   5.1861   5.6061   50.36   <.0001
drug           -0.0098   0.1231  -0.2511   0.2315   -0.08   0.9364
sqrtweek       -0.4048   0.0739  -0.5496  -0.2600   -5.48   <.0001
drug*sqrtweek  -0.5243   0.0840  -0.6890  -0.3596   -6.24   <.0001
Scale           1.2186    .        .        .         .      .
NOTE: The scale parameter for GEE estimation was computed as the square
      root of the normalized Pearson's chi-square.