Table III Variable selection metrics
ModelaModificationsSelection criteria
AICbBICbrfitrLOOCVSNRPSW
1Full model−2653−10160.9920.9899.2<3 × 10−16
2Remove terms−2741−18270.9940.9888.5<3 × 10−16
3Remove terms−2772−19880.9930.9908.5<3 × 10−16
4Remove terms−2772−21130.9930.9908.3<3 × 10−16
5Remove terms−2807−24260.9930.9908.1<3 × 10−16
6Remove terms−2806−24460.9920.9908.1<3 × 10−16
7Remove outliers−2981−26210.9930.9928.72.1 × 10−6
8Remove terms (“Final model”)2983−26280.9930.9928.72.1 × 10−6
  • a Eight iterations of variable selection were performed, each resulting in a different model.

  • b The AIC and BIC metrics were computed from models fit according to the maximum likelihood criterion instead of the restricted maximum likelihood criterion because the former is required for correct AIC and BIC estimates (28).