Since the relationship between birthweight and gestational is not linear, we used fractional polynomial regression to determine the best functional model which resulted in adding four fractional polynomial functions of gestational age to the model.
All continuous variables were examined for the best functional form before model testing using the −2 log-likelihood ratio test and visual inspection of plots.
From this family of models, the best functional form is chosen using Akaike's information criterion (AIC).
The best functional form f 3 of G-ALPHA1 is f 3 ∈ P [ 4 ] (Fig. S3C).
Morphology, especially ecologically-critical morphology, has functional requirements, thus it makes sense that the best performing models of tooth row shape include both kinetic-type in addition to static-type measures.
Results of the models including this outcome and its best functional predictor (the WOMAC score) are shown in Table 6.
To identify the best functional form for the baseline, we compared the models with Akaike Information Criterion AICC).
In the final multivariable model (see Table 5, 7% of variance accounted for), compared to those with best functional status (i.e. "fully active") each level of worsening functional status was associated with worse POS score.
Molecular discovery often involves identification of the best functional groups (substituents) on a scaffold.
Within each group we applied different versions of the model to the data in order to detect the best functional form for describing the relationship between the populations and the nitrogen cycle (for the nitrogen-dependent population growth models).
