Repository files navigation

dcor: distance correlation and energy statistics in Python.

E-statistics are functions of distances between statistical observationsin metric spaces.

Distance covariance and distance correlation aredependency measures between random vectors introduced in[SRB07] witha simple E-statistic estimator.

This package offers functions for calculating several E-statisticssuch as:

• Estimator of the energy distance[SR13].
• Biased and unbiased estimators of distance covariance anddistance correlation[SRB07].
• Estimators of the partial distance covariance and partialdistance covariance[SR14].

It also provides tests based on these E-statistics:

• Test of homogeneity based on the energy distance.
• Test of independence based on distance covariance.

dcor is on PyPi and can be installed usingpip:

pip install dcor

It is also available forconda using theconda-forge channel:

conda install -c conda-forge dcor

Previous versions of the package were in thevnmabus channel. Thischannel will not be updated with new releases, and users are recommended touse theconda-forge channel.

dcor is available in Python 3.8 or above in all operating systems.The package dcor depends on the following libraries:

• numpy
• numba >= 0.51
• scipy
• joblib

Please, if you find this software useful in your work, reference it citing the following paper:

@article{ramos-carreno+torrecilla_2023_dcor,  author = {Ramos-Carreño, Carlos and Torrecilla, José L.},  doi = {10.1016/j.softx.2023.101326},  journal = {SoftwareX},  month = {2},  title = {{dcor: Distance correlation and energy statistics in Python}},  url = {https://www.sciencedirect.com/science/article/pii/S2352711023000225},  volume = {22},  year = {2023},}

You can additionally cite the software repository itself using:

@misc{ramos-carreno_2022_dcor,  author = {Ramos-Carreño, Carlos},  doi = {10.5281/zenodo.3468124},  month = {3},  title = {dcor: distance correlation and energy statistics in Python},  url = {https://github.com/vnmabus/dcor},  year = {2022}}

If you want to reference a particular version for reproducibility, check the version-specific DOIs available in Zenodo.

The documentation can be found inhttps://dcor.readthedocs.io/en/latest/?badge=latest