Diverging Roads: Theory-Based vs. Machine Learning-Implied Stock Risk Premia

69 Pages Posted: 24 Feb 2020 Last revised: 23 Mar 2020

See all articles by Joachim Grammig

Joachim Grammig

Eberhard Karls Universität Tübingen; Centre for Financial Research (CfR); Center for Financial Studies (CfS)

Constantin Hanenberg

University of Tuebingen - Faculty of Economics and Social Sciences

Christian Schlag

Leibniz Institute for Financial Research SAFE

Jantje Sönksen

University of Tübingen

Date Written: February 12, 2020

Abstract

We assess financial theory-based and machine learning-implied measurements of stock risk premia by comparing the quality of their return forecasts. In the low signal-to-noise environment of a one month horizon, we find that it is preferable to rely on a theory-based approach instead of engaging in the computer-intensive hyper-parameter tuning of statistical models. The theory-based approach also delivers a solid performance at the one year horizon, at which only one machine learning methodology (random forest) performs substantially better. We also consider ways to combine the opposing modeling philosophies, and identify the use of random forests to account for the approximation residuals of the theory-based approach as a promising hybrid strategy. It combines the advantages of the two diverging paths in the finance world.

Keywords: stock risk premia, return forecasts, machine learning, theory-based return prediction

JEL Classification: C53, C58, G12, G17

Suggested Citation

Grammig, Joachim and Hanenberg, Constantin and Schlag, Christian and Sönksen, Jantje, Diverging Roads: Theory-Based vs. Machine Learning-Implied Stock Risk Premia (February 12, 2020). Available at SSRN: https://ssrn.com/abstract=3536835 or http://dx.doi.org/10.2139/ssrn.3536835

Joachim Grammig

Eberhard Karls Universität Tübingen ( email )

Mohlstrasse 36
D-72074 Tübingen, 72074
Germany

HOME PAGE: http://www.uni-tuebingen.de/uni/wwo/Grammig/grammigeng.html

Centre for Financial Research (CfR) ( email )

Albertus-Magnus Platz
Cologne, 50923
Germany

Center for Financial Studies (CfS) ( email )

Taunusanlage 6
Frankfurt/Germany, D-60329
Germany

Constantin Hanenberg

University of Tuebingen - Faculty of Economics and Social Sciences ( email )

Mohlstrasse 36
Tuebingen, Baden-Wuerttemberg 72074
Germany

Christian Schlag

Leibniz Institute for Financial Research SAFE ( email )

(http://www.safe-frankfurt.de)
Theodor-W.-Adorno-Platz 3
Frankfurt am Main, 60323
Germany
+49 69 798 33699 (Phone)

Jantje Sönksen (Contact Author)

University of Tübingen ( email )

Sigwartstr. 18
Tübingen, Baden-Wuerttemberg 72076
Germany

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
102
Abstract Views
355
rank
274,502
PlumX Metrics