OPTIONS NOCENTER; TITLE1 'Analysis of Riesby data - HDRS scores across time'; DATA ONE; INFILE 'C:\mixdemo\RIESBY.DAT'; INPUT ID HamD Intcpt Week Endog EndWeek ; PROC FORMAT; VALUE Endog 0='NonEndog' 1='Endog'; VALUE Week 0='week 0' 1='week 1' 2='week 2' 3='week 3' 4='week 4' 5='week 5'; PROC MIXED METHOD=ML COVTEST; CLASS ID; MODEL HAMD = WEEK /SOLUTION; RANDOM INTERCEPT /SUB=ID TYPE=UN G; TITLE2 'Random intercepts model: compound symmetry structure'; PROC MIXED METHOD=ML COVTEST; CLASS ID; MODEL HAMD = WEEK /SOLUTION; RANDOM INTERCEPT WEEK /SUB=ID TYPE=UN G GCORR; TITLE2 'Random trend model'; PROC MIXED METHOD=ML COVTEST; CLASS ID; MODEL HAMD = WEEK ENDOG ENDWEEK /SOLUTION; RANDOM INTERCEPT WEEK /SUB=ID TYPE=UN G GCORR; TITLE2 'Random trend model with group effects'; PROC MIXED METHOD=ML COVTEST; CLASS ID; MODEL HAMD = WEEK WEEK*WEEK /SOLUTION; RANDOM INTERCEPT WEEK WEEK*WEEK /SUB=ID TYPE=UN G GCORR; TITLE2 'Random quadratic trend model'; RUN; ---------- OUTPUT ---------- Analysis of Riesby data - HDRS scores across time Random intercepts model: compound symmetry structure The Mixed Procedure Model Information Data Set WORK.ONE Dependent Variable HamD 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 2 Columns in X 2 Columns in Z Per Subject 1 Subjects 66 Max Obs Per Subject 6 Observations Used 375 Observations Not Used 21 Total Observations 396 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 2399.71226762 1 2 2285.18878668 0.00000000 Convergence criteria met. Estimated G Matrix Row Effect ID Col1 1 Intercept 101 16.1544 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) ID 16.1544 3.4099 4.74 <.0001 Residual 19.0378 1.5316 12.43 <.0001 Fit Statistics -2 Log Likelihood 2285.2 AIC (smaller is better) 2293.2 AICC (smaller is better) 2293.3 BIC (smaller is better) 2301.9 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 1 114.52 <.0001 Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 23.5518 0.6385 65 36.88 <.0001 Week -2.3757 0.1350 308 -17.60 <.0001 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F Week 1 308 309.80 <.0001 Analysis of Riesby data - HDRS scores across time Random trend model The Mixed Procedure Model Information Data Set WORK.ONE Dependent Variable HamD 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 2 Columns in Z Per Subject 2 Subjects 66 Max Obs Per Subject 6 Observations Used 375 Observations Not Used 21 Total Observations 396 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 2399.71226762 1 2 2219.30726751 0.00031526 2 1 2219.04369935 0.00000794 3 1 2219.03751634 0.00000001 Convergence criteria met. Estimated G Matrix Row Effect ID Col1 Col2 1 Intercept 101 12.6280 -1.4197 2 Week 101 -1.4197 2.0779 Estimated G Correlation Matrix Row Effect ID Col1 Col2 1 Intercept 101 1.0000 -0.2771 2 Week 101 -0.2771 1.0000 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) ID 12.6280 3.5272 3.58 0.0002 UN(2,1) ID -1.4197 1.0372 -1.37 0.1711 UN(2,2) ID 2.0779 0.5162 4.03 <.0001 Residual 12.2177 1.1191 10.92 <.0001 Fit Statistics -2 Log Likelihood 2219.0 AIC (smaller is better) 2231.0 AICC (smaller is better) 2231.3 BIC (smaller is better) 2244.2 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 3 180.67 <.0001 Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 23.5769 0.5455 65 43.22 <.0001 Week -2.3771 0.2086 65 -11.39 <.0001 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F Week 1 65 129.84 <.0001 Analysis of Riesby data - HDRS scores across time Random trend model with group effects The Mixed Procedure Model Information Data Set WORK.ONE Dependent Variable HamD 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 6 Observations Used 375 Observations Not Used 21 Total Observations 396 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 2390.43887457 1 2 2215.16683509 0.00028011 2 1 2214.93417137 0.00000635 3 1 2214.92924669 0.00000000 Convergence criteria met. Estimated G Matrix Row Effect ID Col1 Col2 1 Intercept 101 11.6402 -1.4006 2 Week 101 -1.4006 2.0762 Estimated G Correlation Matrix Row Effect ID Col1 Col2 1 Intercept 101 1.0000 -0.2849 2 Week 101 -0.2849 1.0000 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) ID 11.6402 3.3589 3.47 0.0003 UN(2,1) ID -1.4006 1.0157 -1.38 0.1679 UN(2,2) ID 2.0762 0.5160 4.02 <.0001 Residual 12.2193 1.1193 10.92 <.0001 Fit Statistics -2 Log Likelihood 2214.9 AIC (smaller is better) 2230.9 AICC (smaller is better) 2231.3 BIC (smaller is better) 2248.4 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 3 175.51 <.0001 Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 22.4763 0.7943 64 28.30 <.0001 Week -2.3657 0.3118 64 -7.59 <.0001 Endog 1.9880 1.0690 243 1.86 0.0641 EndWeek -0.02706 0.4194 243 -0.06 0.9486 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F Week 1 64 57.58 <.0001 Endog 1 243 3.46 0.0641 EndWeek 1 243 0.00 0.9486 Analysis of Riesby data - HDRS scores across time Random quadratic trend model The Mixed Procedure Model Information Data Set WORK.ONE Dependent Variable HamD 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 7 Columns in X 3 Columns in Z Per Subject 3 Subjects 66 Max Obs Per Subject 6 Observations Used 375 Observations Not Used 21 Total Observations 396 Iteration History Iteration Evaluations -2 Log Like Criterion 0 1 2399.57194989 1 3 2207.80524380 0.00019159 2 1 2207.64984879 0.00000256 3 1 2207.64788628 0.00000000 Convergence criteria met. Estimated G Matrix Row Effect ID Col1 Col2 Col3 1 Intercept 101 10.4400 -0.9151 -0.1122 2 Week 101 -0.9151 6.6379 -0.9365 3 Week*Week 101 -0.1122 -0.9365 0.1937 Estimated G Correlation Matrix Row Effect ID Col1 Col2 Col3 1 Intercept 101 1.0000 -0.1099 -0.07888 2 Week 101 -0.1099 1.0000 -0.8258 3 Week*Week 101 -0.07888 -0.8258 1.0000 Covariance Parameter Estimates Standard Z Cov Parm Subject Estimate Error Value Pr Z UN(1,1) ID 10.4400 3.5862 2.91 0.0018 UN(2,1) ID -0.9151 2.4076 -0.38 0.7039 UN(2,2) ID 6.6379 2.7613 2.40 0.0081 UN(3,1) ID -0.1122 0.4209 -0.27 0.7898 UN(3,2) ID -0.9365 0.4881 -1.92 0.0550 UN(3,3) ID 0.1937 0.09364 2.07 0.0193 Residual 10.5163 1.1061 9.51 <.0001 Fit Statistics -2 Log Likelihood 2207.6 AIC (smaller is better) 2227.6 AICC (smaller is better) 2228.3 BIC (smaller is better) 2249.5 Null Model Likelihood Ratio Test DF Chi-Square Pr > ChiSq 6 191.92 <.0001 Solution for Fixed Effects Standard Effect Estimate Error DF t Value Pr > |t| Intercept 23.7602 0.5521 65 43.04 <.0001 Week -2.6326 0.4790 65 -5.50 <.0001 Week*Week 0.05148 0.08835 65 0.58 0.5621 Type 3 Tests of Fixed Effects Num Den Effect DF DF F Value Pr > F Week 1 65 30.21 <.0001 Week*Week 1 65 0.34 0.5621