Skip to main content

Table 2 Percentage of the features with zero effect size for every rank position

From: A comparison of methods for interpreting random forest models of genetic association in the presence of non-additive interactions

Sample size 1000
Information gain: IG2 IG3
Percent of cases: 25% 50% 25 50%
 F1 0% 0% 0% 0%
 F2 0% 0% 0% 0%
 F3 0% 15% 0% 0%
 F4 76% 70% 23% 83%
 F5 91% 92% 65% 92%
Sample size 10,000
Information gain: IG2 IG3
Percent of cases: 25% 50% 25% 50%
 F1 0% 0% 0% 0%
 F2 0% 0% 0% 0%
 F3 0% 17% 0% 0%
 F4 85% 54% 16% 47%
 F5 94% 84% 59% 88%
  1. F1, F2, etc. – feature ranks, PFI permutation feature importance, BIC build-in coefficients, SHAP shapley additive explanations, IG information gain
\