TITLE1 'partitioning BS and WS effects of drug levels'; DATA one; INFILE 'c:\mixdemo\riesbyt4.dat'; INPUT id hamdelt intcpt week sex endog lnimi lndmi; PROC SORT; BY id; PROC MEANS NOPRINT; CLASS id; VAR lnimi lndmi; OUTPUT OUT = two MEAN = mlnimi mlndmi; DATA three; MERGE one two; BY id; lnidev = lnimi - mlnimi; lnddev = lndmi - mlndmi; PROC MIXED METHOD=ML COVTEST; CLASS id; MODEL hamdelt = week lnimi lndmi /SOLUTION; RANDOM INTERCEPT week /SUB=id TYPE=UN G GCORR; TITLE2 'assuming bs=ws drug effects'; PROC MIXED METHOD=ML COVTEST; CLASS id; MODEL hamdelt = week mlnimi mlndmi lnidev lnddev /SOLUTION; RANDOM INTERCEPT week /SUB=id TYPE=UN G GCORR; TITLE2 'relaxing bs=ws drug effects'; RUN; ----------- OUTPUT ----------- partitioning BS and WS effects of drug levels assuming bs=ws drug effects The Mixed Procedure Model Information Data Set WORK.THREE Dependent Variable hamdelt Covariance Structure Unstructured Subject Effect id Estimation Method ML Residual Variance Method Profile Fixed Effects SE Method Model-Based Degrees of Freedom Method Containment Class Level Information Class Levels Values id 66 101 103 104 105 106 107 108 113 114 115 117 118 120 121 123 302 303 304 305 308 309 310 311 312 313 315 316 318 319 322 327 328 331 333 334 335 337 338 339 344 345 346 347 348 349 350 351 352 353 354 355 357 360 361 501 502 504 505 507 603 604 606 607 608 609 610 Dimensions Covariance Parameters 4 Columns in X 4 Columns in Z Per Subject 2 Subjects 66 Max Obs Per Subject 4 Observations Used 250 Observations Not Used 1 Total Observations 251 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 1635.34280196 1 2 1499.12180567 0.00047011 2 1 1498.85298630 0.00001381 3 1 1498.84565871 0.00000001 4 1 1498.84565143 0.00000000 Convergence criteria met. Estimated G Matrix Row Effect id Col1 Col2 1 Intercept 101 20.4997 0.8372 2 week 101 0.8372 2.7824 Estimated G Correlation Matrix Row Effect id Col1 Col2 1 Intercept 101 1.0000 0.1108 2 week 101 0.1108 1.0000 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) id 20.4997 5.1266 4.00 <.0001 UN(2,1) id 0.8372 1.6144 0.52 0.6041 UN(2,2) id 2.7824 0.9692 2.87 0.0020 Residual 10.5278 1.3852 7.60 <.0001 Fit Statistics -2 Log Likelihood 1498.8 AIC (smaller is better) 1514.8 AICC (smaller is better) 1515.4 BIC (smaller is better) 1532.4 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 3 136.50 <.0001 partitioning BS and WS effects of drug levels assuming bs=ws drug effects The Mixed Procedure Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 1.5214 3.7426 65 0.41 0.6857 week -1.9669 0.2850 65 -6.90 <.0001 lnimi 0.6301 0.8211 116 0.77 0.4444 lndmi -1.9666 0.6025 116 -3.26 0.0014 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F week 1 65 47.64 <.0001 lnimi 1 116 0.59 0.4444 lndmi 1 116 10.66 0.0014 Model Information Data Set WORK.THREE Dependent Variable hamdelt Covariance Structure Unstructured Subject Effect id Estimation Method ML Residual Variance Method Profile Fixed Effects SE Method Model-Based Degrees of Freedom Method Containment Class Level Information Class Levels Values id 66 101 103 104 105 106 107 108 113 114 115 117 118 120 121 123 302 303 304 305 308 309 310 311 312 313 315 316 318 319 322 327 328 331 333 334 335 337 338 339 344 345 346 347 348 349 350 351 352 353 354 355 357 360 361 501 502 504 505 507 603 604 606 607 608 609 610 Dimensions Covariance Parameters 4 Columns in X 6 Columns in Z Per Subject 2 Subjects 66 Max Obs Per Subject 4 Observations Used 250 Observations Not Used 1 Total Observations 251 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 1632.98838927 1 2 1495.93931787 0.00029636 2 1 1495.77268363 0.00000601 3 1 1495.76952679 0.00000000 Convergence criteria met. Estimated G Matrix Row Effect id Col1 Col2 1 Intercept 101 20.3201 0.4979 2 week 101 0.4979 2.8250 Estimated G Correlation Matrix Row Effect id Col1 Col2 1 Intercept 101 1.0000 0.06572 2 week 101 0.06572 1.0000 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) id 20.3201 5.1012 3.98 <.0001 UN(2,1) id 0.4979 1.6441 0.30 0.7620 UN(2,2) id 2.8250 0.9739 2.90 0.0019 Residual 10.3769 1.3591 7.64 <.0001 Fit Statistics -2 Log Likelihood 1495.8 AIC (smaller is better) 1515.8 AICC (smaller is better) 1516.7 BIC (smaller is better) 1537.7 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 3 137.22 <.0001 Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 7.2669 5.0388 64 1.44 0.1541 week -2.0238 0.2917 65 -6.94 <.0001 mlnimi -0.3129 1.0037 115 -0.31 0.7558 mlndmi -2.3671 0.7963 115 -2.97 0.0036 lnidev 2.4434 1.4561 115 1.68 0.0960 lnddev -1.7963 0.9987 115 -1.80 0.0747 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F week 1 65 48.14 <.0001 mlnimi 1 115 0.10 0.7558 mlndmi 1 115 8.84 0.0036 lnidev 1 115 2.82 0.0960 lnddev 1 115 3.23 0.0747