Algorithms for detecting associations, dynamical influences and causal inference from data.
104 Stars
Updated Last
1 Year Ago
Started In
May 2018


CI codecov DOI

CausalityTools.jl is a package for quantifying associations and dynamical coupling between datasets, independence testing and causal inference.

All further information is provided in the documentation, which you can either find online or build locally by running the docs/make.jl file.

Key features

  • Association measures from conventional statistics, information theory and dynamical systems theory, for example distance correlation, mutual information, transfer entropy, convergent cross mapping and a lot more!
  • A dedicated API for independence testing, which comes with automatic compatibility with every measure-estimator combination you can think of. For example, we offer the generic SurrogateTest, which is fully compatible with TimeseriesSurrogates.jl, and the LocalPermutationTest for conditional indepencence testing.
  • A dedicated API for causal network inference based on these measures and independence tests.


To install the package, run import Pkg; Pkg.add("CausalityTools").