MediumTime Seriests-164

Relative Absolute Error to Naive

Problem

Compute the relative absolute error against a naive forecast: sum(|actual - forecast|) / sum(|actual - naive|).

Return 0.0 when the naive denominator is zero.

Hint

Values below 1 mean the forecast beats the naive benchmark.

case 1

in: [[10,12,14],[11,11,11],[10,10,10]]

expected: 0.8333333333333334

case 2

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

expected: 0.6666666666666666

case 3

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

expected: 0

case 4

in: [[1,2],[3,2],[0,2]]

expected: 2

Starter codePython

def relative_absolute_error_naive(actual, forecast, naive):
    # Your code here
    pass

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