Seasonal Naive Baseline MSE

Problem

Compute the MSE of the seasonal naive baseline: the mean of (x[t] - x[t-m])^2 over t >= m.

Return 0.0 if season_length is not positive or the series is not longer than one season.

Hint

A perfectly repeating seasonal pattern gives zero baseline error.

case 1

in: [[1,2,3,4,5,6],2]

expected: 4

case 2

in: [[2,4,6,8],2]

expected: 16

case 3

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

expected: 0

case 4

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

expected: 0

Starter codePython

def seasonal_naive_baseline_mse(series, season_length):
    # Your code here
    pass

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