📦 Announcing `bayesianOU`: A Package for Nonlinear Price Convergence Diagnostics

Package development
, ,
1

Hello R Community,

I'm excited to announce the first stable release of bayesianOU, an R package now available on CRAN. This package is designed for researchers in economic dynamics and finance interested in testing mean-reversion and market price convergence.

What does it do?
Drawing from the wiki documentation, bayesianOU offers a principled econometric framework to test whether observed market prices systematically revert toward theoretical equilibrium values. The core functions help you:

  • Model Nonlinear Dynamics: Estimate Ornstein-Uhlenbeck processes with cubic drift to capture increasing mean-reversion speeds at larger deviations.
  • Handle Complex Market Features: Incorporate stochastic volatility and heavy-tailed (Student-t) innovations to account for real-world market turbulence.
  • Quantify Convergence: Formally test mean-reversion hypotheses with hierarchical Bayesian inference, providing exact posterior probabilities and half-life estimates.

At its core, the package solves the problem of detecting systematic attraction between disaggregated economic variables (e.g., market prices vs. production prices) in the presence of volatility clustering and structural heterogeneity—a task where traditional linear models often fail.

Why is it useful?
Ensuring rigorous convergence testing is vital for validating economic theories of value and price adjustment. bayesianOU moves beyond simple linear regressions, offering a transparent and reproducible Bayesian toolkit that accounts for the nonlinear and stochastic nature of economic data.

Getting Started:

The package vignettes and wiki provide detailed examples, from testing the labor theory of value to analyzing sectoral price dynamics. I hope bayesianOU becomes a valuable tool in your time series or financial econometrics workflow.

I welcome any feedback, bug reports, or contributions on GitHub. What are your experiences or challenges with modeling mean-reversion in volatile markets?

Best,

José Mauricio Gómez Julián