Multiscale Graph Correlation

A Powerful Testing Procedure

Multiscale Graph Correlation (MGC, pronounced "magic") is a nonparametric approach for independence and k-sample testing.

An easy-to-use package

Currently available in Python, with many other standard independence and k-sample tests included. Development versions in both R (github and CRAN) and MATLAB support only MGC.

A scalable solution

Our implementations are fast and powerful.


C. Shen, C. E. Priebe and J. T. Vogelstein. From Distance Correlation to Multiscale Graph Correlation. Journal of the American Statistical Association, 2018.

Y. Lee, C. Shen and J. T. Vogelstein. Network Dependence Testing via Diffusion Maps and Distance-Based Correlations. Biometrika, 2018.