Detecting Sparse Cointegration
Methodology
2026-03-05 v2 Econometrics
Abstract
We propose a two-step procedure to detect cointegration in high-dimensional settings, focusing on sparse relationships. First, we use the adaptive LASSO to identify the small subset of integrated covariates driving the equilibrium relationship with a target series, ensuring model-selection consistency. Second, we adopt an information-theoretic model choice criterion to distinguish between stationarity and nonstationarity in the resulting residuals, avoiding dependence on asymptotic distributional assumptions. Monte Carlo experiments confirm robust finite-sample performance, even under endogeneity and serial correlation.
Cite
@article{arxiv.2501.13839,
title = {Detecting Sparse Cointegration},
author = {Jesus Gonzalo and Jean-Yves Pitarakis},
journal= {arXiv preprint arXiv:2501.13839},
year = {2026}
}