IPW Mean

Problem

Return the inverse probability weighted mean of the treated outcomes, (1/n) * sum(t_i*y_i/e_i) over all n units, where e_i is the propensity score. Return 0.0 when input lengths differ, any treated unit has e_i <= 0, or the input is empty.

Hint

Treated units stand in for themselves and similar untreated units.

Test cases

case 1

in: [[1,0,1,0],[10,5,8,4],[0.5,0.5,0.4,0.6]]

expected: 10

case 2

in: [[0,0],[1,2],[0.2,0.3]]

expected: 0

case 3

in: [[1],[7],[0.7]]

expected: 10

case 4

in: [[],[],[]]

expected: 0

case 5

in: [[1,1],[2,3],[0,0.5]]

expected: 0

Starter codePython

def ipw_mean(treatment, outcome, propensity):
    # Your code here
    pass

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