Feature Selection for Portfolio Optimization

25 Pages Posted: 14 Jan 2015 Last revised: 3 Mar 2017

See all articles by Thomas Bjerring

Thomas Bjerring

Technical University of Denmark - Management Engineering

Omri Ross

University of Copenhagen

Alex Weissensteiner

Free University of Bolzano Bozen

Date Written: January 26, 2016

Abstract

Most portfolio selection rules based on the sample mean and covariance matrix perform poorly out-of-sample. Moreover, there is a growing body of evidence that such optimization rules are not able to beat simple rules of thumb, such as 1/N. Parameter uncertainty has been identified as one major reason for these findings. A strand of literature addresses this problem by improving the parameter estimation and/or by relying on more robust portfolio selection methods. Independent of the chosen portfolio selection rule, we propose using feature selection first in order to reduce the asset menu. While most of the diversification benefits are preserved, the parameter estimation problem is alleviated. We conduct out-of-sample back-tests to show that in most cases different well-established portfolio selection rules applied on the reduced asset universe are able to improve alpha relative to different prominent factor models.

Keywords: Portfolio Optimization, Feature Selection, Agglomerative Hierarchical Clustering

Suggested Citation

Bjerring, Thomas and Ross, Omri and Weissensteiner, Alex, Feature Selection for Portfolio Optimization (January 26, 2016). Annals of Operation Research, Forthcoming. Available at SSRN: https://ssrn.com/abstract=2548800 or http://dx.doi.org/10.2139/ssrn.2548800

Thomas Bjerring

Technical University of Denmark - Management Engineering ( email )

Produktionstorvet 424
room 043
Kgs. Lyngby, 2800
Denmark

Omri Ross (Contact Author)

University of Copenhagen ( email )

Nørregade 10
Copenhagen, København DK-1165
Denmark

Alex Weissensteiner

Free University of Bolzano Bozen ( email )

Universitätsplatz 1
Bolzano, 39100
+39 0471 013496 (Phone)

Register to save articles to
your library

Register

Paper statistics

Downloads
755
Abstract Views
2,901
rank
33,073
PlumX Metrics