Local Predictability in High Dimensions

43 Pages Posted: 31 Jan 2023 Last revised: 13 Jul 2023

See all articles by Philipp Adämmer

Philipp Adämmer

University of Greifswald

Sven Lehmann

University of Rostock - Department of Economics

Rainer Alexander Schüssler

University of Rostock - Department of Economics; University of Konstanz

Date Written: July 5, 2023

Abstract

We propose a novel time series forecasting method designed to handle vast sets of predictive signals, many of which are irrelevant or short-lived. The method transforms heterogeneous scalar-valued signals into candidate density forecasts via time-varying coefficient models, and subsequently, combines them into a final density forecast via time-varying subset combination. Our approach is computationally fast, because it uses online prediction and updating. We validate our method through simulation analyses and apply it to forecast daily aggregate stock returns as well as quarterly inflation, using over 12,000 and over 400 signals, respectively. We find superior forecast performance and lower computation time of our approach relative to competitive benchmark methods.

Keywords: Big Data, Ensemble Learning, Time Series, Stock Returns, Inflation

JEL Classification: C53; C55

Suggested Citation

Adämmer, Philipp and Lehmann, Sven and Schüssler, Rainer Alexander, Local Predictability in High Dimensions (July 5, 2023). Available at SSRN: https://ssrn.com/abstract=4342487 or http://dx.doi.org/10.2139/ssrn.4342487

Philipp Adämmer

University of Greifswald ( email )

Friedrich-Loeffler-Strasse 70
D-17487 Greifswald, 17489
Germany

Sven Lehmann

University of Rostock - Department of Economics ( email )

Ulmenstr. 69
Rostock, 18057
Germany

Rainer Alexander Schüssler (Contact Author)

University of Rostock - Department of Economics ( email )

Ulmenstr. 69
Rostock, 18057
Germany

University of Konstanz ( email )

Fach D-144
Universitätsstraße 10
Konstanz, D-78457
Germany

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
184
Abstract Views
721
Rank
299,485
PlumX Metrics