The Lasso for High-Dimensional Regression with a Possible Change-Point

TL;DR

This paper introduces a Lasso-based method for high-dimensional regression models with potential change-points, enabling simultaneous variable and model selection, with theoretical guarantees and practical validation.

Contribution

It develops a Lasso estimator that jointly selects covariates and determines the presence of a change-point, providing non-asymptotic bounds and handling high-dimensional settings.

Findings

Oracle inequalities for prediction and estimation errors.

Threshold parameter estimation error can be nearly $n^{-1}$.

Method performs well in simulations and real data applications.

Abstract

We consider a high-dimensional regression model with a possible change-point due to a covariate threshold and develop the Lasso estimator of regression coefficients as well as the threshold parameter. Our Lasso estimator not only selects covariates but also selects a model between linear and threshold regression models. Under a sparsity assumption, we derive non-asymptotic oracle inequalities for both the prediction risk and the $\ell_1$ estimation loss for regression coefficients. Since the Lasso estimator selects variables simultaneously, we show that oracle inequalities can be established without pretesting the existence of the threshold effect. Furthermore, we establish conditions under which the estimation error of the unknown threshold parameter can be bounded by a nearly $n^{-1}$ factor even when the number of regressors can be much larger than the sample size ($n$). We illustrate the usefulness of our proposed estimation method via Monte Carlo simulations and an application to real data.