Durbin-Watson Statistic

Problem

Return the Durbin-Watson statistic sum((e_i - e_(i-1))^2) / sum(e_i^2) for a list of residuals. Return 0.0 when fewer than two residuals are given or all residuals are zero.

Hint

Values near 2 indicate no autocorrelation; values near 0 indicate positive autocorrelation.

Test cases

case 1

in: [[1,1,1,1]]

expected: 0

case 2

in: [[1,-1,1,-1]]

expected: 3

case 3

in: [[1,2,1,2]]

expected: 0.3

case 4

in: [[0,0,0]]

expected: 0

case 5

in: [[2]]

expected: 0

Starter codePython

def durbin_watson(residuals):
    # Your code here
    pass

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