Seasonal Naive h-Step Forecast

Problem

Forecast h steps beyond the end of a series with the seasonal naive method of period m. The forecast repeats the value one full season earlier, cycling through the last m observations: the index used is n - m + ((h - 1) mod m).

Return 0.0 when m is not positive, when the series is shorter than one full season, or when h is not positive.

Hint

The forecast pattern wraps around the last complete season.

Test cases

case 1

in: [[10,20,30,40,50,60],2,1]

expected: 50

case 2

in: [[10,20,30,40,50,60],2,2]

expected: 60

case 3

in: [[10,20,30,40,50,60],2,3]

expected: 50

case 4

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

expected: 0

case 5

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

expected: 2

Starter codePython

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

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