Autoregressive covariance for strength data

Exercise Therapy Trial

 

The Mixed Procedure

 

Model Information

Data Set

WORK.EXERCISE

Dependent Variable

y

Covariance Structure

Autoregressive

Subject Effect

id

Estimation Method

REML

Residual Variance Method

Profile

Fixed Effects SE Method

Model-Based

Degrees of Freedom Method

Between-Within

 

Class Level Information

Class

Levels

Values

id

37

not printed

group

2

1 2

time

5

0 4 6 8 12

 

Dimensions

Covariance Parameters

2

Columns in X

18

Columns in Z

0

Subjects

37

Max Obs Per Subject

5

Observations Used

173

Observations Not Used

12

Total Observations

185

 

Iteration History

Iteration

Evaluations

-2 Res Log Like

Criterion

0

1

893.47252907

 

1

2

621.46236120

0.00218463

2

1

621.07958857

0.00004781

3

1

621.07179183

0.00000002

4

1

621.07178807

0.00000000

 

Convergence criteria met.

 

Estimated R Matrix for id 1

Row

Col1

Col2

Col3

Col4

Col5

1

11.8673

11.1573

10.4899

9.8623

9.2723

2

11.1573

11.8673

11.1573

10.4899

9.8623

3

10.4899

11.1573

11.8673

11.1573

10.4899

4

9.8623

10.4899

11.1573

11.8673

11.1573

5

9.2723

9.8623

10.4899

11.1573

11.8673

 

Estimated R Correlation Matrix for id 1

Row

Col1

Col2

Col3

Col4

Col5

1

1.0000

0.9402

0.8839

0.8311

0.7813

2

0.9402

1.0000

0.9402

0.8839

0.8311

3

0.8839

0.9402

1.0000

0.9402

0.8839

4

0.8311

0.8839

0.9402

1.0000

0.9402

5

0.7813

0.8311

0.8839

0.9402

1.0000

 

Covariance Parameter Estimates

Cov Parm

Subject

Estimate

AR(1)

id

0.9402

Residual

 

11.8673

 

Fit Statistics

-2 Res Log Likelihood

621.1

AIC (smaller is better)

625.1

AICC (smaller is better)

625.1

BIC (smaller is better)

628.3

 

 

 

Type 3 Tests of Fixed Effects

Effect

Num DF

Den DF

Chi-Square

F Value

Pr > ChiSq

Pr > F

group

1

35

1.72

1.72

0.1893

0.1979

time

4

128

34.21

8.55

<.0001

<.0001

group*time

4

128

2.94

0.74

0.5679

0.5697