MIDAS-QR with 2-Dimensional Structure

TL;DR

This paper introduces a novel MIDAS-QR model with structured regularization across both lag and quantile dimensions, improving nowcasting and forecasting accuracy for US data.

Contribution

It extends MIDAS-QR by imposing structure on both lag and quantile dimensions, reducing unnecessary variation and enhancing predictive performance.

Findings

Improved forecast accuracy in US data

More gradual lag profiles achieved

Enhanced model stability and interpretability

Abstract

Mixed frequency data has been shown to improve the performance of growth-at-risk models in the literature. Most of the research has focused on imposing structure on the high-frequency lags when estimating MIDAS-QR models akin to what is done in mean models. However, only imposing structure on the lag-dimension can potentially induce quantile variation that would otherwise not be there. In this paper we extend the framework by introducing structure on both the lag dimension and the quantile dimension. In this way we are able to shrink unnecessary quantile variation in the high-frequency variables. This leads to more gradual lag profiles in both dimensions compared to the MIDAS-QR and UMIDAS-QR. We show that this proposed method leads to further gains in nowcasting and forecasting on a pseudo-out-of-sample exercise on US data.