Changepoint Penalty BIC

Problem

Compute the BIC-style penalty for adding a changepoint with p parameters: p * ln(n) for n observations.

Return 0.0 if n or p is not positive.

Hint

The penalty grows with sample size, but only logarithmically.

case 1

in: [100,1]

expected: 4.605170185988092

case 2

in: [100,3]

expected: 13.815510557964275

case 3

in: [1,2]

expected: 0

case 4

in: [10,0]

expected: 0

Starter codePython

import math
def changepoint_penalty_bic(n, p):
    # Your code here
    pass

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