statsmodels.stats.multicomp.pairwise_tukeyhsd

statsmodels.stats.multicomp.pairwise_tukeyhsd(endog, groups, alpha=0.05)[source]

calculate all pairwise comparisons with TukeyHSD confidence intervals

this is just a wrapper around tukeyhsd method of MultiComparison

Parameters:
  • endog (ndarray, float, 1d) – response variable
  • groups (ndarray, 1d) – array with groups, can be string or integers
  • alpha (float) – significance level for the test
Returns:

results – A results class containing relevant data and some post-hoc calculations

Return type:

TukeyHSDResults instance

See also

MultiComparison, tukeyhsd, statsmodels.sandbox.stats.multicomp.TukeyHSDResults