search

snoplusuk / echidna

MIT License
4 stars 12 forks source link

Chi squared bug #127

Closed mjmottram closed 8 years ago

mjmottram commented 8 years ago

Is this a bug: https://github.com/snoplusuk/echidna/blob/master/echidna/limit/chi_squared.py#L189. Presumably if observed is < epsilon then you adjust observed, not expected. I have some optimisations for these functions that I'll add this fix to (if indeed it needs it).

ashleyrback commented 8 years ago

Yes this is a typo, it should be:

if observed[i] < epsilon:
    observed[i] = epsilon
ashleyrback commented 8 years ago

@mjmottram this was fixed in #114.