MediumTime Seriests-119

Rolling Kurtosis

Problem

Compute the rolling population excess kurtosis with window size k: mean((x - window_mean)^4) / variance^2 - 3.

Return an empty list if k is not positive or exceeds the series; windows with zero variance produce 0.0 by convention.

Hint

Excess kurtosis is 0 for a normal distribution and negative for light tails.

case 1

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

expected: [-1.5,-1.5,-1.5]

case 2

in: [[1,1,1,2,10],3]

expected: [0,-1.4999999999999993,-1.4999999999999996]

case 3

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

expected: [-1.5]

case 4

in: [[5,5],2]

expected: [0]

Starter codePython

def rolling_kurtosis(series, k):
    # Your code here
    pass

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