Brunner-Munzel test that two samples have the relative effect P(X < Y) + P(X = Y)/2 = 1/2, without assuming equal
variances or shapes. Equivalent to lawstat::brunner.munzel.test() in R. NaN values are dropped.
estimate is the estimated relative effect and confInt its two-sided interval (not clipped to [0, 1]). Each
sample needs at least 2 values (lawstat returns NaN for a sample of one). When the variance estimates are 0 (for
example completely separated samples), the statistic, degrees of freedom and p-value are NaN or ±Infinity, as in
lawstat.
Brunner-Munzel test that two samples have the relative effect P(X < Y) + P(X = Y)/2 = 1/2, without assuming equal variances or shapes. Equivalent to
lawstat::brunner.munzel.test()in R.NaNvalues are dropped.estimateis the estimated relative effect andconfIntits two-sided interval (not clipped to [0, 1]). Each sample needs at least 2 values (lawstat returns NaN for a sample of one). When the variance estimates are 0 (for example completely separated samples), the statistic, degrees of freedom and p-value areNaNor ±Infinity, as in lawstat.