statsmodels.stats.weightstats._tconfint_generic(mean, std_mean, dof, alpha, alternative)[source]

generic t-confint based on summary statistic

meanfloat or ndarray

Value, for example mean, of the first sample.

std_meanfloat or ndarray

Standard error of the difference value1 - value2

dofint or float

Degrees of freedom


Significance level for the confidence interval, coverage is 1-alpha.


The alternative hypothesis, H1, has to be one of the following

  • ‘two-sided’ : H1: value1 - value2 - diff not equal to 0.

  • ‘larger’ : H1: value1 - value2 - diff > 0

  • ‘smaller’ : H1: value1 - value2 - diff < 0

lowerfloat or ndarray

Lower confidence limit. This is -inf for the one-sided alternative “smaller”.

upperfloat or ndarray

Upper confidence limit. This is inf for the one-sided alternative “larger”.