The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

infometrics

Information-Theoretic Methods for Econometric Estimation

infometrics is an R package implementing Information-Theoretic (IT) estimators for econometric models, following the unified framework of Golan (2008), Information and Entropy Econometrics — A Review and Synthesis (doi:10.1561/0800000004). The package focuses on the Generalized Maximum Entropy (GME) and Generalized Cross-Entropy (GCE) family: each unknown is reparameterized as the mean of a distribution on a bounded support, and the resulting entropy problem is solved subject to the data.

Every estimator solves the dual (concentrated) model — working in Lagrange-multiplier space via BFGS with analytic gradients — which is computationally efficient and needs no compiled code. The package is pure R and depends only on stats.

Installation

# install.packages("remotes")
remotes::install_github("GenMaxEnt/infometrics_R")

What’s included

One function per estimator family: uniform priors give GME, user-supplied priors (p0, w0) give GCE — there are no separate gce_* twins.

Regression

Inverse problems

Matrix balancing and Markov chains

Multinomial response

Utilitiesshannon_entropy(), make_support(), default_supports(), normalize_data(), plus the margins() and fano_bounds() generics.

Every estimator returns an S3 object inheriting from "infometrics" with print, summary, coef, fitted, and residuals methods (and, where applicable, vcov, predict, margins, and fano_bounds).

Vignettes

vignette("inverse-problems",    package = "infometrics")
vignette("gme-gce-regression",  package = "infometrics")

A note on support spaces

Support spaces are the user’s most consequential input. The dual is bounded below only if the data are representable within the supports — if they are not, the multipliers diverge and estimates collapse to a support boundary. The estimators check the first-order condition at the optimum (foc_residual) and warn when it is violated; the remedy is a wider error support v (and/or Z).

Reference

Golan, A. (2008). Information and Entropy Econometrics — A Review and Synthesis. Foundations and Trends in Econometrics, 2(1–2), 1–145. doi:10.1561/0800000004

License

GPL-3.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.