Effective Sample Size from Autocorrelation

Problem

Return the effective sample size n*(1 - rho1)/(1 + rho1) for a series with lag-1 autocorrelation rho1. Return 0.0 when n <= 0 or rho1 is not in (-1, 1].

Hint

Positive autocorrelation reduces the information each observation carries.

Test cases

case 1

in: [100,0.5]

expected: 33.333333333333336

case 2

in: [100,0]

expected: 100

case 3

in: [100,-0.5]

expected: 300

case 4

in: [50,1]

expected: 0

case 5

in: [10,-1]

expected: 0

Starter codePython

def effective_sample_size(n, rho1):
    # Your code here
    pass

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