Deep parametric portfolio policies

Arbeitspapier

We directly optimize portfolio weights as a function of firm characteristics via deep neural networks by generalizing the parametric portfolio policy framework. Our results show that network-based portfolio policies result in an increase of investor utility of between 30 and 100 percent over a comparable linear portfolio policy, depending on whether portfolio restrictions on individual stock weights, short-selling or transaction costs are imposed, and depending on an investor's utility function. We provide extensive model interpretation and show that network-based policies better capture the non-linear relationship between investor utility and firm characteristics. Improvements can be traced to both variable interactions and non-linearity in functional form. Both the linear and the network-based approach agree on the same dominant predictors, namely past return-based firm characteristics.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: CFR Working Paper ; No. 23-01

Classification and Topics

• Classification: Wirtschaft ; Portfolio Choice; Investment Decisions ; Asset Pricing; Trading Volume; Bond Interest Rates ; Financial Econometrics ; Neural Networks and Related Topics
• Subject: Portfolio Choice ; Machine Learning ; Expected Utility

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Simon, Frederik ; Weibels, Sebastian ; Zimmermann, Tom
• Event: Veröffentlichung
• (who): University of Cologne, Centre for Financial Research (CFR)
• (where): Cologne
• (when): 2023

Further information

• Handle: http://hdl.handle.net/10419/270745
• Last update: 10.03.2025, 11:43 AM CET