Forecast Bias Significance

Problem

Test whether forecast errors have a significant mean bias with a t-style statistic: mean(errors) / (std(errors) / sqrt(n)), using the population standard deviation.

Return 0.0 when there are fewer than two errors or the standard deviation is zero.

Hint

A large positive value indicates a systematic tendency to under-forecast.

Test cases

case 1

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

expected: 0

case 2

in: [[1,2,3]]

expected: 4.242640687119285

case 3

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

expected: 0

case 4

in: [[0.5,1.5,1]]

expected: 4.242640687119285

Starter codePython

def forecast_bias_significance(errors):
    # Your code here
    pass

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