Skip to content

Instantly share code, notes, and snippets.

@hobson
Created July 13, 2015 01:15
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
Star You must be signed in to star a gist
Save hobson/7ec4091bec74f6157824 to your computer and use it in GitHub Desktop.
"""Attempt to confirm results in paper by Miller and Sanjurjo
["Confused by the Gambler's and Hot Hand Falacy?..."](http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2627354)
Calculate the probability of a streak of length 2 in a coin flip
for a small sequence of attempts ("finite sequence of iid Bernoulli trials").
The fencepost counting issue means you have to ignore one of the flips.
Perhaps this is what causes Miller and Sanjurjo to get 0.4 instead of 0.5.
"the conditional relative frequency... is 0.4 not 0.5"
"""
from random import randint
h, t = 0, 0
# L is the streak length but is unused (2 is hard coded into the logic)
L, M, N = 2, 1000, 4
for i in range(M):
flip0 = randint(0, 1)
for j in range(N-1):
flip1 = randint(0, 1)
if flip0:
h += flip1
t += not flip1
flip0 = flip1
print float(h) / float(h + t)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment