statsmodels.stats.weightstats.DescrStatsW.ttost_mean

DescrStatsW.ttost_mean(low, upp)[source]

test of (non-)equivalence of one sample

TOST: two one-sided t tests

null hypothesis: m < low or m > upp alternative hypothesis: low < m < upp

where m is the expected value of the sample (mean of the population).

If the pvalue is smaller than a threshold, say 0.05, then we reject the hypothesis that the expected value of the sample (mean of the population) is outside of the interval given by thresholds low and upp.

Parameters:
low, uppfloat

equivalence interval low < mean < upp

Returns:
pvaluefloat

pvalue of the non-equivalence test

t1, pv1, df1tuple

test statistic, pvalue and degrees of freedom for lower threshold test

t2, pv2, df2tuple

test statistic, pvalue and degrees of freedom for upper threshold test