X-Learner Step

Problem

Combine X-learner imputed effects with a weight g for the control model: tau = g * d0_hat + (1 - g) * d1_hat, element-wise.

Empty input returns [].

Hint

The weight g is often the propensity score, favoring the better-estimated arm.

case 1

in: [[1,2],[3,4],0.25]

expected: [2.5,3.5]

case 2

in: [[0,1],[2,-1],0.5]

expected: [1,0]

case 3

in: [[],[],0.5]

expected: []

Starter codePython

def x_learner_step(d0_hat, d1_hat, g):
    # Your code here
    pass

3 test cases · run your code against them in the editor