SIR CI

Problem

Return the log-based confidence interval for a standardized incidence ratio: SIR * exp(-z/sqrt(observed)) to SIR * exp(z/sqrt(observed)). Return [0.0, 0.0] when observed or expected is not positive.

Hint

The standard error of ln(SIR) is 1/sqrt(observed events).

Test cases

case 1

in: [120,100,1.96]

expected: [1.0034046373317664,1.4351139574451428]

case 2

in: [10,8,1.96]

expected: [0.6725607277616321,2.323210284965938]

case 3

in: [5,10,1.645]

expected: [0.2395930199746716,1.0434360734984205]

case 4

in: [0,10,1.96]

expected: [0,0]

case 5

in: [10,0,1.96]

expected: [0,0]

Starter codePython

import math
def sir_ci(observed, expected, z_crit):
    # Your code here
    pass

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