Generalized Black-Litterman with Decision Fusion

EFMA 2023

59 Pages Posted: 30 Mar 2023 Last revised: 17 Apr 2023

See all articles by Xinyu Huang

Xinyu Huang

University of Bath - School of Management

Massimo Guidolin

Bocconi University, Dept. of Finance; Bocconi University - CAREFIN - Centre for Applied Research in Finance

David Newton

University of Bath - School of Management

Emmanouil Platanakis

University of Bath - School of Management

Xiaoxia Ye

University of Exeter Business School - Department of Finance

Date Written: April 16, 2023

Abstract

At the heart of the portfolio selection problem lies the challenge of accurately estimating asset expected returns and covariance matrices. The classical Black-Litterman model addresses this challenge by combining market equilibrium and investors' views within the Markowitz mean-variance framework. This paper identifies three potential weaknesses of this allocation framework and proposes solutions to address them. First, we characterize the equilibrium model as a flexible, data-driven framework and present a new portfolio combination strategy for estimating equilibrium expected returns. Second, we leverage information from multiple sources to understand asset return dynamics comprehensively. Then, we establish a connection between investors' knowledge and the decision-making process through the use of an entropy-based decision-fusion system. Finally, we propose a GLASSO-Wishart model for analyzing realized covariance matrices of asset returns and show how estimation risk can be reduced by restricting the matrix of a Wishart distribution to sparse parameterization. In the presence of transaction costs, the generalized Black-Litterman (GBL) strategy can consistently and significantly increase the Sharpe ratio over the 1/N strategy. Regarding risk-reward maximization, we observe our GBL strategy to outperform eight well-studied benchmark strategies proposed in the literature to manage estimation risk. Extensive robustness experiments show that our approach withstand various choices of parameters.

Keywords: Finance, Investment Analysis, Black-Litterman, Parameter Uncertainty, Decision Fusion

JEL Classification: G11, G17

Suggested Citation

Huang, Xinyu and Guidolin, Massimo and Newton, David and Platanakis, Emmanouil and Ye, Xiaoxia, Generalized Black-Litterman with Decision Fusion (April 16, 2023). EFMA 2023, Available at SSRN: https://ssrn.com/abstract=4395771 or http://dx.doi.org/10.2139/ssrn.4395771

Xinyu Huang

University of Bath - School of Management ( email )

Claverton Down
Bath, BA2 7AY
United Kingdom

Massimo Guidolin

Bocconi University, Dept. of Finance ( email )

Via Roentgen, 1
2nd floor
Milan, MI 20136
Italy

Bocconi University - CAREFIN - Centre for Applied Research in Finance

Via Sarfatti 25
Milan, 20136
Italy

David Newton

University of Bath - School of Management ( email )

Claverton Down
Bath, BA2 7AY
United Kingdom

Emmanouil Platanakis (Contact Author)

University of Bath - School of Management ( email )

Claverton Down
Bath, BA2 7AY
United Kingdom

Xiaoxia Ye

University of Exeter Business School - Department of Finance ( email )

Streatham Court
Exeter, EX4 4PU
United Kingdom

HOME PAGE: http://www.xiaoxiaye.me/

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
250
Abstract Views
824
Rank
223,996
PlumX Metrics