Skip to main content

Table 2 Simulation design for Scenario 1

From: Intervention in prediction measure: a new approach to assessing variable importance for random forests

Variables X 1 X 2 X 3 X 4 X 5 Y|X 2=0 Y|X 2=1
Distribution N(0,1) B(1,0.5) DU(1/4) DU(1/10) DU(1/20) B(1,0.5 - rel) B(1,0.5 + rel)
  1. The variables are sampled independently from the following distributions. N(0,1) stands for the standard normal distribution. B(1, π) stands for the Binomial distribution with n = 1, i.e the Bernoulli distribution, and probability π. DU(1/n) stands for the Discrete Uniform distribution with values 1, …, n. The relevance parameter rel indicates the degree of dependence between Y and X 2, and is set at 0.1, which is not very high