Win Rate Interval Half Width

Problem

The half width of a normal-approximation interval for a win rate is z * sqrt(p (1 - p) / n).

Given the win rate, the number of comparisons, and the critical z, return the half width.

Hint

It is the standard error scaled by the critical value.

case 1

in: [0.6,100,1.96]

expected: 0.09601999791710057

case 2

in: [0.5,400,1.645]

expected: 0.041125

case 3

in: [0.9,50,2.576]

expected: 0.10929042410019278

Starter codePython

def win_rate_interval_half_width(win_rate, n, z):
    # Your code here
    pass

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