Machine Learning Time Series Regressions with an Application to Nowcasting

TL;DR

This paper develops structured machine learning regressions for high-dimensional time series, demonstrating that sparse-group LASSO effectively leverages data structure and improves nowcasting accuracy, especially when incorporating text data.

Contribution

It introduces a sparse-group LASSO approach tailored for high-dimensional, multi-frequency time series, with theoretical guarantees and empirical validation in economic nowcasting.

Findings

Sparse-group LASSO outperforms unstructured LASSO in time series regression.

The method effectively incorporates heavy-tailed data and mixed frequencies.

Including text data enhances nowcasting performance.

Abstract

This paper introduces structured machine learning regressions for high-dimensional time series data potentially sampled at different frequencies. The sparse-group LASSO estimator can take advantage of such time series data structures and outperforms the unstructured LASSO. We establish oracle inequalities for the sparse-group LASSO estimator within a framework that allows for the mixing processes and recognizes that the financial and the macroeconomic data may have heavier than exponential tails. An empirical application to nowcasting US GDP growth indicates that the estimator performs favorably compared to other alternatives and that text data can be a useful addition to more traditional numerical data.