Table 3.

Statistical criteria of the predictive potential for the quantitative structure–property activity relationships (QSPR/QSAR) models.

Criterion of the Predictive Potential	Reference
$R=\frac{\mathrm{n}{\displaystyle \sum }\mathrm{xy}-{\displaystyle \sum }x{\displaystyle \sum }y}{\sqrt{(n{\displaystyle \sum }{x}^2-{({\displaystyle \sum }x)}^2(n{\displaystyle \sum }{y}^2-{({\displaystyle \sum }y)}^2}}$	[75]
$CCC=\frac{2{\displaystyle \sum }\left(\mathrm{x}-\overline{\mathrm{x}}\right)\left(\mathrm{y}-\overline{\mathrm{y}}\right)}{{\displaystyle \sum }\left(\mathrm{x}-\overline{\mathrm{x}}\right){}^2+{\displaystyle \sum }\left(\mathrm{y}-\overline{\mathrm{y}}\right){}^2+\mathrm{n}\left(\overline{\mathrm{x}}-\overline{\mathrm{y}}\right){}^2}$	[76]
$R_0^2=1-\frac{{\displaystyle \sum }{({\tilde{y}}_i-y_i^{ro})}^2}{{\displaystyle \sum }{({\tilde{y}}_i-\overline{\tilde{y}})}^2}$ ${R^{\prime }}_0^2=1-\frac{{\displaystyle \sum }{(y_i-{\tilde{y}}_i^{ro})}^2}{{\displaystyle \sum }{(y_i-\overline{y_i})}^2}$ $k=\frac{y_{i }\widetilde{y_i}}{{\tilde{y}}_i^2}$ $k^{\prime }=\frac{y_{i }\widetilde{y_i}}{y_i^2}$	[77]
$Q^2=1-\frac{{\displaystyle \sum }(y_k-{ý}_k){ }^2}{{\displaystyle \sum }(y_k-{\overline{y}}_k^{ }){ }^2}$ $Q_{F1}^2=1-\frac{\left[{{\displaystyle \sum }}_{i=1}^{N_{EXT}}\left({\acute{y}}_i-y_i\right){ }^2\right]/N_{EXT}}{\left[{{\displaystyle \sum }}_{i=1}^{N_{EXT}}\left(y_i-\overline{y}{ }_{TR}\right){ }^2\right]/N_{EXT}}$ $Q_{F2}^2=1-\frac{\left[{{\displaystyle \sum }}_{i=1}^{N_{EXT}}\left({\acute{y}}_i-y_i\right){ }^2\right]/N_{EXT}}{\left[{{\displaystyle \sum }}_{i=1}^{N_{EXT}}\left(y_i-\overline{y}{ }_{EXT}\right){ }^2\right]/N_{EXT}}$ $Q_{F3}^2=1-\frac{\left[{{\displaystyle \sum }}_{i=1}^{N_{EXT}}\left({\acute{y}}_i-y_i\right){ }^2\right]/N_{EXT}}{\left[{{\displaystyle \sum }}_{i=1}^{N_{TR}}\left(y_i-\overline{y}{ }_{TR}\right){ }^2\right]/N_{TR}}$	[78]
$r_m^2=r^2\left( 1-\sqrt{\left|r^2-r_0^2\right| }\right)$ $\overline{R_m^2}=\frac{R_m^2\left(x,y\right)-R_m^2\left(y,x\right)}{2}$ $\Delta R_m^2=\left|R_m^2\left(x,y\right)-R_m^2\left(y,x\right)\right|$	[79]
$II{C}_{CLB}=r_{CLB}\frac{\min ({}_{ }{}^{-}MAE{ }_{CLB},{}_{ }{}^{+}MAE{ }_{CLB})}{max({}_{ }{}^{-}MAE{ }_{CLB},{}_{ }{}^{+}MAE{ }_{CLB})}$ ${}_{ }{}^{-}MAE{ }_{CLB}=\frac{1}{{}_{ }{}^{-}N}{\displaystyle \sum }_{k=1}^{{}_{ }{}^{-}N}\left| {\Delta }_k\right|,$${\Delta }_{k}0;{}_{ }{}^{-}N is the number of {\Delta }_k<0$ ${}_{ }{}^{+}MAE{ }_{CLB}=\frac{1}{{}_{ }{}^{+}N}{\displaystyle \sum }_{k=1}^{{}_{ }{}^{+}N}\left| {\Delta }_k\right|$, ${\Delta }_{k}0;{}_{ }{}^{+}N is the number of {\Delta }_k\ge 0$ ${\Delta }_k=observe{d}_k-calculate{d}_k$	[80]