Information-Theoretic Measure of Causality

Methods for quantifying temporal and spatial causality through information flow, and decomposing it into unique, redundant, and synergistic components, following the framework described in Martinez-Sanchez et al. (2024) .

infocausality

Information-Theoretic Measure of Causality

infocausality is an R package for information-theoretic causal analysis.
It quantifies temporal and spatial causality through information flow, and decomposes it into unique, redundant, and synergistic components. The package provides native support for data.frame, sf, and SpatRaster objects, offering a unified interface for both time-series and spatial cross-sectional causal analysis.

Refer to the package documentation https://stscl.github.io/infocausality/ for more detailed information.

⚠️ Note: The SURD (Synergistic-Unique-Redundant Decomposition) core computations in infocausality are executed via Python bindings. A pure C++ implementation with improved performance and easier deployment has been developed in the infoxtr package. For new projects, we recommend using infoxtr as a drop-in replacement.

Installation

• Install from CRAN with:

install.packages("infocausality", dep = TRUE)

• Install binary version from R-universe with:

install.packages("infocausality",
                 repos = c("https://stscl.r-universe.dev",
                           "https://cloud.r-project.org"),
                 dep = TRUE)

• Install from source code on GitHub with:

if (!requireNamespace("devtools")) {
    install.packages("devtools")
}
devtools::install_github("stscl/infocausality",
                         build_vignettes = TRUE,
                         dep = TRUE)