Table 2.

Partition model’s performance using % predicted change in FEV₁

Measure	Training	Validation	Definition
Entropy R2	0.715	0.715	${\displaystyle 1-\mathrm{L}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}(\mathrm{m}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{l})/\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}(0)}$
Generalized R2	0.934	0.934	${\displaystyle (1-(\mathrm{L}(0)/\mathrm{L}(\mathrm{m}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{l}){)}^{\wedge }(2/\mathrm{n}))/(1-\mathrm{L}(0{)}^{\wedge }(2/\mathrm{n}))}$
Mean −log p	0.446	0.445	${\displaystyle \sum -\mathrm{L}\mathrm{o}\mathrm{g}(\rho [\mathrm{j}])/\mathrm{n}}$
RASE	0.381	0.382	${\displaystyle \sqrt{}\sum (\mathrm{y}[\mathrm{j}]-\rho [\mathrm{j}]{)}^2/\mathrm{n}}$
Mean Abs Dev	0.268	0.268	${\displaystyle \sum \left|\mathrm{y}[\mathrm{j}]-\rho [\mathrm{j}]\right|/\mathrm{n}}$
Misclassification rate	0.20	0.21	${\displaystyle \sum (\rho [\mathrm{j}]\ne \rho \mathrm{M}\mathrm{a}\mathrm{x})/\mathrm{n}}$
N	17 543	7, 4, 68	n

FEV₁, forced expiratory volume in 1 s; Mean Abs Dev, mean absolute deviance; RASE, square root of the mean squared prediction error.