CUPED Adjusted Estimate

Problem

CUPED adjusts a treatment effect estimate using a pre-experiment covariate: y_adjusted = mean(y) - theta * (mean(x) - mean(x_prior)).

Given the post-experiment mean, the covariate mean, the pre-period covariate mean, and theta, return the adjusted estimate.

Hint

Shift the mean by the scaled covariate imbalance.

case 1

in: [1,5,4,0.5]

expected: 0.5

case 2

in: [10,10,10,1]

expected: 10

case 3

in: [3,8,6,1.5]

expected: 0

Starter codePython

def cuped_adjusted_estimate(mean_y, mean_x, mean_x_prior, theta):
    # Your code here
    pass

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