Naive Forecast MAPE

Problem

Compute the mean absolute percentage error of the one-step naive forecast over the last h observations, where each forecast is the previous observation. Each error is 100 * |actual - forecast| / |actual|.

Observations equal to zero are skipped because the percentage error is undefined. Return 0.0 when no valid observation remains or when h is outside the range 1 to n - 1.

Hint

MAPE is undefined at zero actuals, so those points cannot contribute an error.

Test cases

case 1

in: [[100,110,99],2]

expected: 10.1010101010101

case 2

in: [[0,5],1]

expected: 100

case 3

in: [[0,0],1]

expected: 0

case 4

in: [[1,0,5],2]

expected: 100

case 5

in: [[1,2,3],5]

expected: 0

Starter codePython

def naive_mape(series, h):
    # Your code here
    pass

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