Skip to main content

Table 2 Prediction error for random forest analyses

From: Two-stage approach for identifying single-nucleotide polymorphisms associated with rheumatoid arthritis using random forests and Bayesian networks

  ITbp ITtop50 IT0  
Statistics Training Test Training Test Training Test CaMML Test
Mean 11.28 14.05 12.73 13.60 14.60 14.73 12.42
SD 0.83 0.90 0.95 0.85 0.96 0.91 0.97
Min 9.80 12.20 10.93 11.60 12.27 12.20 10.35
Max 14.73 16.87 16.00 15.47 18.00 16.53 16.00
p-Valuea 5.26 × 10-18   1.35 × 10-9   0.25   
Difference in median -2.77   -0.93   -0.17   
  1. ap-Value of the paired Wilcoxon rank test comparing training and test data prediction error.