7.1 Introduction

The generalized random forest is a method that is quite flexible in estimating the quantity of interest. The theory of it is built using the moment criterion:

\(E[\psi_{\theta_i, \; \upsilon_i} (O_i) | X_i] = 0, \; for \; all \; x \; in \; \chi\)

Getting down to the nuts and bolts of the theory is beyond the scope of this write-up. Rather, we would want to take a closer look at causal forests – a component of GRF framework.