Double Exponential Smoothing

Problem

Apply Holt's double exponential smoothing and return [level, trend]. Initialize level = x[0] and trend = x[1] - x[0], then for each later x[t] compute new_level = alpha * x[t] + (1 - alpha) * (level + trend) and new_trend = beta * (new_level - level) + (1 - beta) * trend.

A single-element series returns [x[0], 0.0] and an empty series returns [0.0, 0.0].

Hint

Update the level first, then use the fresh level to update the trend.

Test cases

case 1

in: [[1,2,3,4,5],0.5,0.5]

expected: [5,1]

case 2

in: [[10,12,14,16],0.3,0.2]

expected: [16,2]

case 3

in: [[10],0.5,0.5]

expected: [10,0]

case 4

in: [[2,4],0.5,0.5]

expected: [4,2]

Starter codePython

def double_exponential_smoothing(series, alpha, beta):
    # Your code here
    pass

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