The SAS System 10:06 Sunday, July 13, 2008 1 The Mixed Procedure Model Information Data Set WORK.FEV1UNI Dependent Variable fev1 Covariance Structure Unstructured Subject Effect patient(drug) Estimation Method REML Residual Variance Method None Fixed Effects SE Method Model-Based Degrees of Freedom Method Between-Within Class Level Information Class Levels Values drug 3 a c p patient 24 201 202 203 204 205 206 207 208 209 210 211 212 214 215 216 217 218 219 220 221 222 223 224 232 hour 8 1 2 3 4 5 6 7 8 Dimensions Covariance Parameters 36 Columns in X 36 Columns in Z 0 Subjects 72 Max Obs Per Subject 8 Number of Observations Number of Observations Read 576 Number of Observations Used 576 Number of Observations Not Used 0 Iteration History Iteration Evaluations -2 Res Log Like Criterion 0 1 1263.95663130 1 1 197.46436713 0.00000000 Convergence criteria met. The SAS System 10:06 Sunday, July 13, 2008 2 The Mixed Procedure Estimated R Matrix for patient(drug) 201 a Row Col1 Col2 Col3 Col4 Col5 Col6 Col7 1 0.4541 0.4587 0.4441 0.4154 0.4349 0.3934 0.3562 2 0.4587 0.5163 0.4808 0.4688 0.4943 0.4254 0.3992 3 0.4441 0.4808 0.4923 0.4687 0.4843 0.4263 0.4021 4 0.4154 0.4688 0.4687 0.4938 0.4837 0.4179 0.4023 5 0.4349 0.4943 0.4843 0.4837 0.5779 0.4945 0.4643 6 0.3934 0.4254 0.4263 0.4179 0.4945 0.4906 0.4454 7 0.3562 0.3992 0.4021 0.4023 0.4643 0.4454 0.4994 8 0.3840 0.4257 0.4256 0.4251 0.4950 0.4632 0.4496 Estimated R Matrix for patient(drug) 201 a Row Col8 1 0.3840 2 0.4257 3 0.4256 4 0.4251 5 0.4950 6 0.4632 7 0.4496 8 0.5031 Estimated R Correlation Matrix for patient(drug) 201 a Row Col1 Col2 Col3 Col4 Col5 Col6 Col7 1 1.0000 0.9474 0.9393 0.8773 0.8490 0.8334 0.7479 2 0.9474 1.0000 0.9536 0.9284 0.9050 0.8453 0.7861 3 0.9393 0.9536 1.0000 0.9506 0.9080 0.8674 0.8109 4 0.8773 0.9284 0.9506 1.0000 0.9055 0.8490 0.8102 Estimated R Correlation Matrix for patient(drug) 201 a Row Col8 1 0.8033 2 0.8352 3 0.8552 4 0.8529 The SAS System 10:06 Sunday, July 13, 2008 3 The Mixed Procedure Estimated R Correlation Matrix for patient(drug) 201 a Row Col1 Col2 Col3 Col4 Col5 Col6 Col7 5 0.8490 0.9050 0.9080 0.9055 1.0000 0.9286 0.8642 6 0.8334 0.8453 0.8674 0.8490 0.9286 1.0000 0.8997 7 0.7479 0.7861 0.8109 0.8102 0.8642 0.8997 1.0000 8 0.8033 0.8352 0.8552 0.8529 0.9181 0.9323 0.8970 Estimated R Correlation Matrix for patient(drug) 201 a Row Col8 5 0.9181 6 0.9323 7 0.8970 8 1.0000 Covariance Parameter Estimates Cov Parm Subject Estimate UN(1,1) patient(drug) 0.4541 UN(2,1) patient(drug) 0.4587 UN(2,2) patient(drug) 0.5163 UN(3,1) patient(drug) 0.4441 UN(3,2) patient(drug) 0.4808 UN(3,3) patient(drug) 0.4923 UN(4,1) patient(drug) 0.4154 UN(4,2) patient(drug) 0.4688 UN(4,3) patient(drug) 0.4687 UN(4,4) patient(drug) 0.4938 UN(5,1) patient(drug) 0.4349 UN(5,2) patient(drug) 0.4943 UN(5,3) patient(drug) 0.4843 UN(5,4) patient(drug) 0.4837 UN(5,5) patient(drug) 0.5779 UN(6,1) patient(drug) 0.3934 UN(6,2) patient(drug) 0.4254 UN(6,3) patient(drug) 0.4263 UN(6,4) patient(drug) 0.4179 UN(6,5) patient(drug) 0.4945 UN(6,6) patient(drug) 0.4906 UN(7,1) patient(drug) 0.3562 UN(7,2) patient(drug) 0.3992 UN(7,3) patient(drug) 0.4021 UN(7,4) patient(drug) 0.4023 The SAS System 10:06 Sunday, July 13, 2008 4 The Mixed Procedure Covariance Parameter Estimates Cov Parm Subject Estimate UN(7,5) patient(drug) 0.4643 UN(7,6) patient(drug) 0.4454 UN(7,7) patient(drug) 0.4994 UN(8,1) patient(drug) 0.3840 UN(8,2) patient(drug) 0.4257 UN(8,3) patient(drug) 0.4256 UN(8,4) patient(drug) 0.4251 UN(8,5) patient(drug) 0.4950 UN(8,6) patient(drug) 0.4632 UN(8,7) patient(drug) 0.4496 UN(8,8) patient(drug) 0.5031 Fit Statistics -2 Res Log Likelihood 197.5 AIC (smaller is better) 269.5 AICC (smaller is better) 274.6 BIC (smaller is better) 351.4 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 35 1066.49 <.0001 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F drug 2 69 3.60 0.0327 hour 7 69 13.72 <.0001 drug*hour 14 69 4.06 <.0001