Statistical tests on xarray objects. xstatstests currently includes the following statistical tests (most are currently simple, often vectorized, wrappers of scipy.stats functions):
xstatstests.ks_1samp_1d: one-dimensional Kolmogorov-Smirnov test on one samplexstatstests.ks_2samp_1d: one-dimensional Kolmogorov-Smirnov test on two independent samplesxstatstests.ks_2samp_2d: two-dimensional Kolmogorov-Smirnov test on two independent samplesxstatstests.anderson_ksamp: Anderson-Darling test on K independent samplesxstatstests.cramervonmises: Cramér-von Mises test on one samplexstatstests.cramervonmises_2samp: Cramér-von Mises test on two independent samplesxstatstests.epps_singleton_2samp: Epps-Singleton test on two independent samplesxstatstests.ansari: Ansari-Bradley test for equal distribution scale parameters from two independent samplesxstatstests.mood: Mood test for equal distribution scale parameters from two independent samples
xstatstests.ttest_ind: t-test for the means of two independent samplesxstatstests.ttest_rel: t-test for the means of two related samplesxstatstests.bartlett: Bartlett test for the variances of K independent samplesxstatstests.levene: Levene test for the variances of K independent samplesxstatstests.fligner: Fligner-Killeen test for the variances of K independent samplesxstatstests.median_test: Mood test for the medians of K independent samples
xstatstests.skewtest: D’Agostino test for normal sample skewnessxstatstests.kurtosistest: Anscombe test for normal sample kurtosisxstatstests.normaltest: D’Agostino/Pearson test for normal sample skewness and kurtosisxstatstests.jarque_bera: Jarque-Bera test for normal sample skewness and kurtosisxstatstests.shapiro: Shapiro-Wilk test for normality
xstatstests.mannwhitneyu: Mann-Whitney U rank test on two independent samplesxstatstests.ranksums: Wilcoxon rank-sum statistic on two independent samplesxstatstests.kruskal: Kruskal-Wallis H-test on K independent samplesxstatstests.friedmanchisquare: Friedman chi-squared test on K repeated samplesxstatstests.brunnermunzel: Brunner-Munzel test on two independent samples
This package was originally called xks but was renamed when tests additional to the KS test were added.
To install this package from PyPI:
pip install xstatstests
Contributions are very welcome, particularly in the form of reporting bugs and writing tests. Please open an issue and check out the contributor guide.
Press, W.H. et al. 2007, Numerical Recipes, section 14.8
Peacock, J.A. 1983, Two-Dimensional Goodness-of-Fit Testing in Astronomy, Monthly Notices of the Royal Astronomical Society, vol. 202, pp. 615-627
Fasano, G. and Franceschini, A. 1987, A Multidimensional Version of the Kolmogorov-Smirnov Test, Monthly Notices of the Royal Astronomical Society, vol. 225, pp. 155-170