KOMA - Large Macroeconomic Model

koma is an R package for Bayesian estimation of simultaneous equation models (SEMs) using Metropolis-within-Gibbs Markov Chain Monte Carlo (MCMC) methods.

⚠️ Beta Release: koma is under active development. The user-facing API may still change before the first CRAN release. Core features are working but exported functions may change and bugs are possible. Please report issues on GitHub.

Installation

Install the released version from CRAN:

install.packages("koma")

Or install the latest development version from GitHub:

# Install from GitHub
devtools::install_github("timothymerlin/koma")

# Or using renv
renv::install("github::timothymerlin/koma", rebuild = TRUE)

Documentation

• Getting started → Getting started with koma
• Equation syntax → Equation syntax reference
• Extended time series → Extended time series (ets)
• Parallelization → Executing koma in parallel
• Example: Klein model → Estimating Klein’s Model I
• Example: small macro model → Estimating small macro model for Switzerland

Development

Creating a new Version

Increment the package version using usethis:

# Increment the version number (patch/minor/major)
usethis::use_version()

This updates the DESCRIPTION file.

Commit changes and tag the release

After updating the version:

# Stage all changes
git add .

# Commit with a clear message
git commit -m "Incrementing version to x.x.x"

# Tag the release (protected)
git tag x.x.x

# Tag the latest release (unprotected, movable)
git tag -f latest

• x.x.x is the new semantic version (e.g. 0.1.0).
• Protected tags (like x.x.x) are immutable to ensure release integrity.
• latest is an uprotected tag that can be moved to point to the newest release.

Push branch and tags

# Push commits
git push

# Push both tags
git push origin --tags