Table 1 Model Checking
From: Consequence of War on Diabetic Mellitus patients in Tigray region, Ethiopia: a longitudinal study
1.1. Checking for model extension | 1.2. Comparison of models to check whether random effect(s) are needed. | 1.3. Comparison of Models to select the best on for parameter estimation. | |||||
|---|---|---|---|---|---|---|---|
Model | Log likelihood | Model | AIC | Log Likelihood(LL) | |||
Model | F-value | P-value | 1: With no random effect | -58558.95 | Generalized Linear Mixed Model (GLMM) | 107060.0 | -53519.0 |
Linear | 24.012 | <0.0001 | 2: With only random intercept | -58105.2 | Nonlinear Mixed Model (NLMM) | 116587 | -58287.5 |
Quadratic | 65.0527 | <0.0001 | 3: With random intercept and random slope | -56624.1 | |||
Cubic | 53.9426 | <0.0001 | 4: Random intercept and quadratic slope | -54935.45 | |||
Quartic | 41.1079 | 0.0787 | 5: With random intercept and cubic slope | -53518.95 | |||