# statsmodels.stats.weightstats.ztost¶

statsmodels.stats.weightstats.ztost(x1, low, upp, x2=None, usevar='pooled', ddof=1.0)[source]

Equivalence test based on normal distribution

Parameters
x1array_like

one sample or first sample for 2 independent samples

low, uppfloat

equivalence interval low < m1 - m2 < upp

x1

second sample for 2 independent samples test. If None, then a one-sample test is performed.

usevarstr, ‘pooled’

If pooled, then the standard deviation of the samples is assumed to be the same. Only pooled is currently implemented.

Returns
pvaluefloat

pvalue of the non-equivalence test

t1, pv1tuple of floats

test statistic and pvalue for lower threshold test

t2, pv2tuple of floats

test statistic and pvalue for upper threshold test

Notes

checked only for 1 sample case