The SAS System

The GLIMMIX Procedure

Model Information
Data Set WORK.DAT_TWISK_L
Response Variable AVAL
Response Distribution Gaussian
Link Function Identity
Variance Function Default
Variance Matrix Blocked By USUBJID
Estimation Technique Restricted Maximum Likelihood
Degrees of Freedom Method Satterthwaite

Class Level Information
Class Levels Values
USUBJID 152 not printed
AVISIT 3 Week 12 Week 24 Week 0
ABLFL 2 Y N
TRT01P 2 TRT CNT

Number of Observations Read 456
Number of Observations Used 388

Dimensions
R-side Cov. Parameters 6
Columns in X 9
Columns in Z per Subject 0
Subjects (Blocks in V) 152
Max Obs per Subject 3

Optimization Information
Optimization Technique Dual Quasi-Newton
Parameters in Optimization 6
Lower Boundaries 3
Upper Boundaries 0
Fixed Effects Profiled
Starting From Data

Iteration History
Iteration Restarts Evaluations Objective
Function
Change Max
Gradient
0 0 4 2997.0773126 . 0.067493
1 0 4 2996.9517757 0.12553694 0.015671
2 0 2 2996.944912 0.00686364 0.021804
3 0 4 2996.9263444 0.01856760 0.005397
4 0 2 2996.9238906 0.00245383 0.00318
5 0 2 2996.9234905 0.00040011 0.004639
6 0 4 2996.9223977 0.00109276 0.002143
7 0 2 2996.9205342 0.00186348 0.000823
8 0 3 2996.9201344 0.00039989 0.000595
9 0 3 2996.9200545 0.00007982 0.000159
10 0 3 2996.9200533 0.00000126 6.256E-6

Convergence criterion (GCONV=1E-8) satisfied.

Fit Statistics
-2 Res Log Likelihood 2996.92
AIC (smaller is better) 3008.92
AICC (smaller is better) 3009.14
BIC (smaller is better) 3027.06
CAIC (smaller is better) 3033.06
HQIC (smaller is better) 3016.29
Generalized Chi-Square 384.00
Gener. Chi-Square / DF 1.00

Covariance Parameter Estimates
Cov Parm Subject Estimate Standard
Error
UN(1,1) USUBJID 233.91 28.2679
UN(2,1) USUBJID 153.18 23.4459
UN(2,2) USUBJID 212.73 26.4548
UN(3,1) USUBJID 131.68 22.0681
UN(3,2) USUBJID 130.97 21.2591
UN(3,3) USUBJID 183.75 23.8790

Solutions for Fixed Effects
Effect ABLFL TRT01P Estimate Standard
Error
DF t Value Pr > |t|
Intercept     128.43 1.5545 133.2 82.62 <.0001
TRT01P   TRT -6.0720 2.2227 132.7 -2.73 0.0072
TRT01P   CNT 0 . . . .
ABLFL Y   1.8412 1.3007 124.1 1.42 0.1594
ABLFL N   0 . . . .
ABLFL*TRT01P Y TRT 2.1836 1.8624 123.5 1.17 0.2433
ABLFL*TRT01P Y CNT 0 . . . .
ABLFL*TRT01P N TRT 0 . . . .
ABLFL*TRT01P N CNT 0 . . . .

Type III Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
TRT01P 1 135.7 5.03 0.0265
ABLFL 1 123.5 9.92 0.0021
ABLFL*TRT01P 1 123.5 1.37 0.2433

ABLFL*TRT01P Least Squares Means
ABLFL TRT01P Estimate Standard
Error
DF t Value Pr > |t| Alpha Lower Upper
Y TRT 126.38 1.8426 136.7 68.59 <.0001 0.05 122.74 130.03
Y CNT 130.27 1.8034 136.8 72.24 <.0001 0.05 126.71 133.84
N TRT 122.36 1.5887 132.3 77.02 <.0001 0.05 119.22 125.50
N CNT 128.43 1.5545 133.2 82.62 <.0001 0.05 125.36 131.51

Differences of ABLFL*TRT01P Least Squares Means
ABLFL TRT01P _ABLFL _TRT01P Estimate Standard Error DF t Value Pr > |t| Alpha Lower Upper
Y TRT Y CNT -3.8884 2.5782 136.7 -1.51 0.1338 0.05 -8.9867 1.2100
Y TRT N TRT 4.0248 1.3330 123 3.02 0.0031 0.05 1.3862 6.6634
Y TRT N CNT -2.0472 2.4107 179.3 -0.85 0.3969 0.05 -6.8042 2.7099
Y CNT N TRT 7.9132 2.4034 179.7 3.29 0.0012 0.05 3.1707 12.6556
Y CNT N CNT 1.8412 1.3007 124.1 1.42 0.1594 0.05 -0.7332 4.4156
N TRT N CNT -6.0720 2.2227 132.7 -2.73 0.0072 0.05 -10.4685 -1.6754

Simple Effect Comparisons of ABLFL*TRT01P Least Squares Means By ABLFL
Simple Effect
Level
TRT01P _TRT01P Estimate Standard
Error
DF t Value Pr > |t| Alpha Lower Upper
ABLFL Y TRT CNT -3.8884 2.5782 136.7 -1.51 0.1338 0.05 -8.9867 1.2100
ABLFL N TRT CNT -6.0720 2.2227 132.7 -2.73 0.0072 0.05 -10.4685 -1.6754