The Surprising Robustness of Partial Least Squares

TL;DR

This paper demonstrates that Partial Least Squares (PLS) is a robust and effective method for high-dimensional forecasting, outperforming traditional techniques like OLS, LASSO, and ridge regression, especially during recent years.

Contribution

The study shows that PLS can outperform standard regression methods in forecasting GDP growth, highlighting its robustness and effectiveness in high-dimensional, limited-observation settings.

Findings

PLS outperforms OLS, LASSO, and ridge in forecasting GDP from 2020 onward.

All methods perform similarly from 2000-2019, validating PLS as a regularization technique.

PLS reduces out-of-sample forecasting error through dimension reduction.

Abstract

Partial least squares (PLS) is a simple factorisation method that works well with high dimensional problems in which the number of observations is limited given the number of independent variables. In this article, we show that PLS can perform better than ordinary least squares (OLS), least absolute shrinkage and selection operator (LASSO) and ridge regression in forecasting quarterly gross domestic product (GDP) growth, covering the period from 2000 to 2023. In fact, through dimension reduction, PLS proved to be effective in lowering the out-of-sample forecasting error, specially since 2020. For the period 2000-2019, the four methods produce similar results, suggesting that PLS is a valid regularisation technique like LASSO or ridge.