Scaled Error Aggregation

Problem

Compute the mean absolute error divided by a user-supplied scale: mean(abs(actual - forecast)) / scale.

Return 0.0 when actual is empty or the scale is zero.

Hint

The scale is usually the in-sample error of a naive benchmark.

case 1

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

expected: 1

case 2

in: [[10,20],[12,18],2]

expected: 1

case 3

in: [[5],[5],3]

expected: 0

case 4

in: [[1,2],[3,4],0]

expected: 0

Starter codePython

def scaled_error_aggregation(actual, forecast, scale):
    # Your code here
    pass

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