Given the substantial increases in precision obtained using k-NN with an optimised value for k, further results were compared using this non-parametric method to parametric models for prediction of the 300 Index.
The abilities of hCGres and K for predicting MTX resistance were compared using ROC curves.
Survival curves for ApcMin/+, K- ras+/+ and ApcMin/+, K- rastmΔ4A/tmΔ4A mice were compared using the Mantel-Haenszel test [ 23].
Means of Kd, Bmax, IC50 were compared using the t-test.
The K n /K s distributions, obtained using model 2c for pairs of groups with different ecology, were compared using the parametric t-test.
Cumulative frequency distributions were compared using the K-S test.
The parametric data were compared using two-sided t test with the Bonferroni correction (alpha1 = alpha/k) for multiple comparisons (k = 5 for intraoperative comparisons and k = 3 for postoperative comparisons).
Averaged data were compared using t-test.
Raters were compared using kappa statistics.
Cumulative frequency distributions were compared using the Kolmogorov Smirnov test (K S test).
