Entropy Estimators for Discrete Random Variables with R Interface

The 'entropy' package by Jean Hausser implements multiple estimators for entropy, mutual information, and related quantities. It includes the shrinkage estimator by Hausser and Strimmer (2009), maximum likelihood, Millow-Madow, Bayesian, and Chao-Shen estimators, and provides an R interface to the NSB estimator. The package also offers functions for estimating Kullback-Leibler divergence, chi-squared divergence, and for discretizing continuous random variables.

Use Cases

• Estimating entropy of discrete random variables based on the implemented shrinkage, maximum likelihood, and Bayesian estimators.
• Calculating mutual information and Kullback-Leibler divergence between variables based on the provided functions.
• Testing for independence using the chi-squared statistic and G statistic with p-values as described.
• Discretizing continuous random variables for information-theoretic analysis using the included functions.

Strengths

• Implements multiple established estimators, including the Hausser and Strimmer (2009) shrinkage estimator.
• Provides an R interface to the NSB estimator, extending its accessibility.
• Offers functions for a range of related quantities like divergence and independence statistics.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Jean Hausser via paperswithcode.

Collection Method

Software package implementation of statistical methods.

Time Range

Temporal coverage is unknown.

Freshness

Last updated date is unknown.

Geography

Spatial coverage is unknown.