In the original paper of Random Forest, the author introduces a concept of strength and correlation. In the appendix section, a step by step guide to calculating these values are given for the bootstrapping case. I have few doubts in this section.

  1. What does $sd(\theta)$ signify? (equation A2).
  2. As suggested by this paper, how the variance of margin is the correlation of two trees? From RF paper the equation of correlation is given by

$$\bar{\rho} = \frac{var(margin)}{(E_{\theta}(sd(\theta)))^2}$$

where $\theta$ is a tree in a random forest.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.