initialization of ECount parameter in Kendall's Tau about giddy HOT 3 CLOSED

Christensen starts the loop with 0 that effectively sets ECount=1, whereas in _calc the loop start directly with 1, while ECount=0

Example:
from scipy.stats import kendalltau
x1 = [0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1]
x2 = [1, 4, 7, 1, 1, 3, 1, 6, 7, 7, 3, 6, 2, 7, 2, 8]
kt12 = Tau(x1,x2)
kt21 = Tau(x2,x1)
tau = kendalltau(x1,x2)[0]

tau
Out[277]: 0.15494494670022804

kt12.tau
Out[278]: 0.2541027354465522

kt21.tau
Out[279]: 0.16666666666666666

from giddy.

Thank you for submitting this issue.

Can you post an example that reveals that this is a bug?

As it stands, we have followed the algorithm reported by Christensen (2005) which initializes ECount=0.

from giddy.

Thanks for the update.

from giddy.