bayesplain.core.dirichlet_multinomial.log_bayes_factor_independence¶
-
bayesplain.core.dirichlet_multinomial.log_bayes_factor_independence(table, concentration=
1.0, scheme='independent_multinomial', fixed='rows')[source]¶ Log Bayes factor for association against independence.
- Parameters:¶
- table : array_like¶
Table of counts, shape
(R, C).- concentration : float or array_like, default 1.0¶
Dirichlet concentration
a.a = 1makes every combination of parameter values equally likely a priori; larger values pull the alternative’s predictions toward the null, soa = 10behaves like a flat prior already updated with nine hypothetical observations per cell. Forscheme='independent_multinomial'a per-column vector is also accepted, which is how aBeta(a, b)prior on a rate is carried into the Bayes factor unchanged.- scheme : str, default 'independent_multinomial'¶
One of
SCHEMES. Pick the one matching what the study design fixed in advance; see the module docstring.- fixed : {'rows', 'columns'}, default 'rows'¶
Which margin was fixed, for the independent-multinomial scheme only. Rows are the groups you sampled.
- Returns:¶
log(BF10). Positive values favour association between rows and columns; negative values favour independence.- Return type:¶
Notes
Exact – gamma functions only, with no integration. Validated against the published values in Jamil et al. (2017); see
bayesplain/tests/test_dirichlet_multinomial.py.