Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/270745 
Year of Publication: 
2023
Series/Report no.: 
CFR Working Paper No. 23-01
Publisher: 
University of Cologne, Centre for Financial Research (CFR), Cologne
Abstract: 
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.
Subjects: 
Portfolio Choice
Machine Learning
Expected Utility
JEL: 
G11
G12
C58
C45
Document Type: 
Working Paper

Files in This Item:
File
Size
856.62 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.