Drift Forecast at Horizon

Problem

Produce a drift forecast h steps beyond the end of a series. The drift is the average first difference, and the forecast is x[-1] + h * drift.

Return 0.0 for an empty series or a negative horizon, and return the last value when the series has a single observation.

Hint

With a perfectly linear series the drift forecast extends the line exactly.

case 1

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

expected: 9

case 2

in: [[10,20,30],1]

expected: 40

case 3

in: [[5],3]

expected: 5

case 4

in: [[],2]

expected: 0

Starter codePython

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

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