Trigg Tracking Signal

Problem

Compute Trigg's tracking signal from a list of forecast errors. Maintain a smoothed error and a smoothed absolute error with the same alpha: E = alpha * e + (1 - alpha) * E and M = alpha * |e| + (1 - alpha) * M, then return E / M.

Return 0.0 for empty input or when the smoothed absolute error is zero.

Hint

Smoothing both the error and its magnitude makes the signal less jumpy than a plain ratio.

Test cases

case 1

in: [[1,1,1],0.5]

expected: 1

case 2

in: [[1,-1,1,-1],0.5]

expected: -0.3333333333333333

case 3

in: [[0,0,0],0.5]

expected: 0

case 4

in: [[],0.3]

expected: 0

case 5

in: [[2,4,6],0.25]

expected: 1

Starter codePython

def trigg_tracking_signal(errors, alpha):
    # Your code here
    pass

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