On LASSO for Predictive Regression

TL;DR

This paper investigates the limitations of LASSO methods in predictive regression with various predictor types and introduces TAlasso, a novel adaptive LASSO variant, to improve variable selection and estimation consistency.

Contribution

The paper reveals that adaptive LASSO cannot eliminate cointegrating variables with zero coefficients and proposes TAlasso to achieve oracle properties in heterogeneous predictor systems.

Findings

TAlasso restores variable selection consistency.

Conventional LASSO fails in coefficient estimation and screening.

Application to S&P 500 returns demonstrates practical effectiveness.

Abstract

Explanatory variables in a predictive regression typically exhibit low signal strength and various degrees of persistence. Variable selection in such a context is of great importance. In this paper, we explore the pitfalls and possibilities of the LASSO methods in this predictive regression framework. In the presence of stationary, local unit root, and cointegrated predictors, we show that the adaptive LASSO cannot asymptotically eliminate all cointegrating variables with zero regression coefficients. This new finding motivates a novel post-selection adaptive LASSO, which we call the twin adaptive LASSO (TAlasso), to restore variable selection consistency. Accommodating the system of heterogeneous regressors, TAlasso achieves the well-known oracle property. In contrast, conventional LASSO fails to attain coefficient estimation consistency and variable screening in all components simultaneously. We apply these LASSO methods to evaluate the short- and long-horizon predictability of S\&P 500 excess returns.