'Adaptive Seriational Risk Parity' and other Extensions for Heuristic Portfolio Construction using Machine Learning and Graph Theory

Peter Schwendner, Jochen Papenbrock, Markus Jaeger and Stephan Krügel The Journal of Financial Data Science Fall 2021, jfds.2021.1.078; DOI: https://doi.org/10.3905/jfds.2021.1.078

Posted: 18 Mar 2021 Last revised: 7 Oct 2021

See all articles by Peter Schwendner

Peter Schwendner

Zurich University of Applied Sciences

Jochen Papenbrock

NVIDIA GmbH

Markus Jaeger

Munich Reinsurance Company, Financial Solutions

Stephan Krügel

Munich Reinsurance Company, Financial Solutions

Date Written: March 17, 2021

Abstract

In this article, the authors present a conceptual framework named 'Adaptive Seriational Risk Parity' (ASRP) to extend Hierarchical Risk Parity (HRP) as an asset allocation heuristic. The first step of HRP (quasi-diagonalization) determining the hierarchy of assets is required for the actual allocation in the second step of HRP (recursive bisectioning). In the original HRP scheme, this hierarchy is found using the single-linkage hierarchical clustering of the correlation matrix, which is a static tree-based method. The authors of this paper compare the performance of the standard HRP with other static and also adaptive tree-based methods, but also seriation-based methods that do not rely on trees. Seriation is a broader concept allowing to reorder the rows or columns of a matrix to best express similarities between the elements. Each discussed variation leads to a different time series reflecting portfolio performance using a 20-year backtest of a multi-asset futures universe. An unsupervised representation learning based on this time series data creates a taxonomy that groups the strategies in high correspondence to the structure of the various types of ASRP. The performance analysis of the variations shows that most of the static tree-based alternatives of HRP outperform the single linkage clustering used in HRP on a risk-adjusted basis. Adaptive tree methods show mixed results and most generic seriation-based approaches underperform.

Keywords: Hierarchical Risk Parity, portfolio allocation, hierarchical structure, seriation

JEL Classification: C15, G11, G17, G0, G1, G2, G15, G24, E44

Suggested Citation

Schwendner, Peter and Papenbrock, Jochen and Jaeger, Markus and Krügel, Stephan, 'Adaptive Seriational Risk Parity' and other Extensions for Heuristic Portfolio Construction using Machine Learning and Graph Theory (March 17, 2021). Peter Schwendner, Jochen Papenbrock, Markus Jaeger and Stephan Krügel The Journal of Financial Data Science Fall 2021, jfds.2021.1.078; DOI: https://doi.org/10.3905/jfds.2021.1.078, Available at SSRN: https://ssrn.com/abstract=3806714 or http://dx.doi.org/10.2139/ssrn.3806714

Peter Schwendner (Contact Author)

Zurich University of Applied Sciences ( email )

School of Management and Law
Technoparkstrasse 2
Winterthur, CH 8401
Switzerland
8400 (Fax)

Jochen Papenbrock

NVIDIA GmbH ( email )

Germany
+49-(0)1741435555 (Phone)

HOME PAGE: http://www.nvidia.com/en-us/industries/finance/

Markus Jaeger

Munich Reinsurance Company, Financial Solutions ( email )

Königinstr. 107
Munich, 80802
Germany

Stephan Krügel

Munich Reinsurance Company, Financial Solutions ( email )

Königinstr. 107
Munich, 80802
Germany

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
1,366
PlumX Metrics