Directional data is ubiquitous in science. Due to its circular nature such data cannot be analyzed with commonly used statistical techniques. Despite the rapid development of specialized methods for directional statistics over the last fifty years, there is only little software available that makes such methods easy to use for practitioners. Over the last years, I have put together a Matlab and Python toolbox covering most important aspects of directional statistics.
- Descriptive statistics: mean, median, dispersion, skewness, kurtosis
- Inferential statistics: nonuniformity tests (parametric and non-parametric), test for mean/median direction, one- and two-factor ANOVA (multisample test), test for symmetry, test for equal concentration parameter, ks-test
- Measures of association: circular-circular, circular-linear
- Von-Mises distribution: random number generation, parameter estimation, pdf
The toolbox is described in detail in the paper:
P. Berens, CircStat: A Matlab Toolbox for Circular Statistics, Journal of Statistical Software, Volume 31, Issue 10, 2009
Please cite this paper, when you use the toolbox for your research, as others have before.
I am always happy to get feedback which helps me to improve the toolbox. Ideally, bug reports or requests are submitted as issues on github, pull requests with bug fixes are also welcome.
Many thanks to Fabian Sinz for a large part of the work with the Python version of the toolbox.