Breslow-Day Homogeneity Lite

Problem

Return a two-stratum Breslow-Day homogeneity statistic (ln OR1 - ln OR2)^2 / (se1^2 + se2^2), where se_i^2 = 1/a + 1/b + 1/c + 1/d for each 2x2 table. Return 0.0 when any cell is zero.

Hint

Large values suggest the stratum-specific odds ratios are not homogeneous.

Test cases

case 1

in: [[[10,20],[30,40]],[[5,15],[10,30]]]

expected: 0.27024978722164184

case 2

in: [[[20,80],[10,90]],[[30,70],[15,85]]]

expected: 0.019464733639750542

case 3

in: [[[10,10],[10,10]],[[5,5],[5,5]]]

expected: 0

case 4

in: [[[5,0],[1,1]],[[1,1],[1,1]]]

expected: 0

case 5

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

expected: 0

Starter codePython

import math
def breslow_day_lite(t1, t2):
    # Your code here
    pass

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