scipy.stats.wilcoxon

scipy.stats.wilcoxon(x, y=None, zero_method='wilcox', correction=False, alternative='two-sided')

Calculate the Wilcoxon signed-rank test.

The Wilcoxon signed-rank test tests the null hypothesis that two related paired samples come from the same distribution. In particular, it tests whether the distribution of the differences x - y is symmetric about zero. It is a non-parametric version of the paired T-test.

Parameters

xarray_like

Either the first set of measurements (in which case y is the second set of measurements), or the differences between two sets of measurements (in which case y is not to be specified.) Must be one-dimensional.

yarray_like, optional

Either the second set of measurements (if x is the first set of measurements), or not specified (if x is the differences between two sets of measurements.) Must be one-dimensional.

zero_method{“pratt”, “wilcox”, “zsplit”}, optional. Default is “wilcox”.

“pratt”:

includes zero-differences in the ranking process, but drops the ranks of the zeros, see [4], (more conservative)

“wilcox”:

discards all zero-differences, the default

“zsplit”:

includes zero-differences in the ranking process and split the zero rank between positive and negative ones

correctionbool, optional

If True, apply continuity correction by adjusting the Wilcoxon rank statistic by 0.5 towards the mean value when computing the z-statistic. Default is False.

alternative{“two-sided”, “greater”, “less”}, optional

The alternative hypothesis to be tested, see Notes. Default is “two-sided”.

Returns

statisticfloat

If alternative is “two-sided”, the sum of the ranks of the differences above or below zero, whichever is smaller. Otherwise the sum of the ranks of the differences above zero.

pvaluefloat

The p-value for the test depending on alternative.

See also

kruskal, mannwhitneyu

Notes

The test has been introduced in [4]. Given n independent samples (xi, yi) from a bivariate distribution (i.e. paired samples), it computes the differences di = xi - yi. One assumption of the test is that the differences are symmetric, see [2]. The two-sided test has the null hypothesis that the median of the differences is zero against the alternative that it is different from zero. The one-sided test has the null hypothesis that the median is positive against the alternative that it is negative (alternative == 'less'), or vice versa (alternative == 'greater.').

The test uses a normal approximation to derive the p-value (if zero_method == 'pratt', the approximation is adjusted as in [5]). A typical rule is to require that n > 20 ([2], p. 383). For smaller n, exact tables can be used to find critical values.

References

1

https://en.wikipedia.org/wiki/Wilcoxon_signed-rank_test

2(1,2,3,4)

Conover, W.J., Practical Nonparametric Statistics, 1971.

3

Pratt, J.W., Remarks on Zeros and Ties in the Wilcoxon Signed Rank Procedures, Journal of the American Statistical Association, Vol. 54, 1959, pp. 655-667. DOI:10.1080/01621459.1959.10501526

4(1,2,3,4)

Wilcoxon, F., Individual Comparisons by Ranking Methods, Biometrics Bulletin, Vol. 1, 1945, pp. 80-83. DOI:10.2307/3001968

5(1,2)

Cureton, E.E., The Normal Approximation to the Signed-Rank Sampling Distribution When Zero Differences are Present, Journal of the American Statistical Association, Vol. 62, 1967, pp. 1068-1069. DOI:10.1080/01621459.1967.10500917

Examples

In [4], the differences in height between cross- and self-fertilized corn plants is given as follows:

>>> d = [6, 8, 14, 16, 23, 24, 28, 29, 41, -48, 49, 56, 60, -67, 75]

Cross-fertilized plants appear to be be higher. To test the null hypothesis that there is no height difference, we can apply the two-sided test:

>>> from scipy.stats import wilcoxon
>>> w, p = wilcoxon(d)
>>> w, p
(24.0, 0.04088813291185591)

Hence, we would reject the null hypothesis at a confidence level of 5%, concluding that there is a difference in height between the groups. To confirm that the median of the differences can be assumed to be positive, we use:

>>> w, p = wilcoxon(d, alternative='greater')
>>> w, p
(96.0, 0.020444066455927955)

This shows that the null hypothesis that the median is negative can be rejected at a confidence level of 5% in favor of the alternative that the median is greater than zero. The p-value based on the approximation is within the range of 0.019 and 0.054 given in [2]. Note that the statistic changed to 96 in the one-sided case (the sum of ranks of positive differences) whereas it is 24 in the two-sided case (the minimum of sum of ranks above and below zero).