Machine Learning Time Series Regressions with an Application to Nowcasting
This provides an incremental improvement for economists and analysts working with financial and macroeconomic forecasting by offering a more efficient method for handling mixed-frequency time series data.
The paper tackles the problem of analyzing high-dimensional time series data with varying frequencies by introducing a structured machine learning regression method using sparse-group LASSO, which outperforms unstructured LASSO. In an empirical application to nowcasting US GDP growth, the estimator performs favorably compared to alternatives and shows that text data can enhance traditional numerical data.
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.