Persistence Baseline RMSE

Problem

Compute the RMSE of the persistence (last value) baseline: the square root of the mean squared one-step change (x[t] - x[t-1])^2.

Return 0.0 for a series with fewer than two points. This baseline is the bar every forecasting model must clear.

Hint

Squaring before averaging penalizes large one-step jumps.

case 1

in: [[1,3,6,10]]

expected: 3.1091263510296048

case 2

in: [[5,5,5]]

expected: 0

case 3

in: [[10,7,9]]

expected: 2.5495097567963922

case 4

in: [[4]]

expected: 0

Starter codePython

def persistence_baseline_rmse(series):
    # Your code here
    pass

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