TABLE 2.
Hurst Versus Distance Model Comparisons.
Anatomical parcellation | Randomised parcellation | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Discovery | Validation | Complete | Discovery | Validation | Complete | |||||||
Fitting | R2 | AIC | R2 | AIC | R2 | AIC | R2 | AIC | R2 | AIC | R2 | AIC |
Linear | 0.02 | −9.70 | 0.056 | −10.48 | 0.004 | −11.15 | 0.02 | −10.80 | 0.00 | −10.55 | 0.005 | −12.04 |
Quadratic | 0.07 | −9.77 | 0.085 | −10.51 | 0.057 | −11.23 | 0.12 | −10.99 | 0.02 | −10.56 | 0.099 | −12.20 |
Cubic | 0.06 | −9.72 | 0.084 | −10.48 | 0.052 | −11.19 | 0.11 | −10.92 | 0.02 | −10.52 | 0.091 | −12.15 |
Exponential | 0.02 | −9.70 | 0.055 | −10.48 | 0.004 | −11.15 | 0.02 | −10.80 | 0.00 | −10.55 | 0.005 | −12.04 |
Exponential + Intersect | 0.02 | −9.67 | 0.055 | −10.45 | 0.003 | −11.12 | 0.02 | −10.76 | 0.00 | −10.51 | 0.001 | −12.00 |
The best fitting model in terms of maximal AIC is highlighted in bold typeface. R2 = coefficient of determination, AIC = Akaike Information Criteria.