Mixed Model Representation of Penalized Spline for Log C-Peptide Concentration
Study of Log C-Peptide Concentration in Children with Diabetes

The Mixed Procedure

Model Information
Data Set WORK.DIABETES
Dependent Variable logc
Covariance Structure Banded Toeplitz
Estimation Method REML
Residual Variance Method Profile
Fixed Effects SE Method Model-Based
Degrees of Freedom Method Containment

Dimensions
Covariance Parameters 2
Columns in X 2
Columns in Z 10
Subjects 1
Max Obs Per Subject 43

Number of Observations
Number of Observations Read 43
Number of Observations Used 43
Number of Observations Not Used 0

Iteration History
Iteration Evaluations -2 Res Log Like Criterion
0 1 -101.62650626  
1 2 -107.15085960 0.00000011
2 1 -107.15087002 0.00000000

Convergence criteria met.

Covariance Parameter Estimates
Cov Parm Estimate Standard Error Z Value Pr > Z
Variance 0.000104 0.000128 0.81 0.2084
Residual 0.003069 0.000699 4.39 <.0001

Fit Statistics
-2 Res Log Likelihood -107.2
AIC (smaller is better) -103.2
AICC (smaller is better) -102.8
BIC (smaller is better) -107.2

Null Model Likelihood Ratio Test
DF Chi-Square Pr > ChiSq
1 5.52 0.0188

Solution for Fixed Effects
Effect Estimate Standard Error DF t Value Pr > |t|
Intercept 0.5246 0.03173 31 16.53 <.0001
age 0.02689 0.006929 31 3.88 0.0005

Solution for Random Effects
Effect Estimate Std Err Pred DF t Value Pr > |t|
bf1 -0.00773 0.008908 31 -0.87 0.3923
bf2 -0.00820 0.009147 31 -0.90 0.3768
bf3 -0.00629 0.009229 31 -0.68 0.5006
bf4 -0.00387 0.009086 31 -0.43 0.6734
bf5 -0.00157 0.009155 31 -0.17 0.8646
bf6 -0.00115 0.009112 31 -0.13 0.9005
bf7 -0.00046 0.009014 31 -0.05 0.9594
bf8 0.000616 0.009261 31 0.07 0.9474
bf9 0.001050 0.009486 31 0.11 0.9125
bf10 0.001376 0.009884 31 0.14 0.8902

Type 3 Tests of Fixed Effects
Effect Num DF Den DF F Value Pr > F
age 1 31 15.06 0.0005



Plot of Pred by age