Does Extreme Correlation Matter in Global Equity Asset Allocation?

25 Pages Posted: 27 Feb 2018

See all articles by Bruno Solnik

Bruno Solnik

Hong Kong University of Science & Technology (HKUST) - Department of Finance ; HEC Paris - Departement Finance et Economie

Thaisiri Watewai

Chulalongkorn University - Department of Banking & Finance

Date Written: February 16, 2018

Abstract

Global asset allocation provides risk diversification. But international market correlation increases sharply during global crises and diversification benefit disappears when it is most needed. We model these correlation breaks and derive the asset allocation implications. The model can quickly detect crises and suggests adapting allocation for changing correlation and volatility, as the crisis probability evolves. The out-of-sample results for ten major equity markets over 2008-2016 show significant improvements in the Sharpe ratio and maximum drawdown over mean-variance, fat-tail distribution, passive indices and 1/N rule. A benefit of the model is that it is conceptually intuitive and amenable to simple implementation in asset allocation and risk management.

Keywords: Extreme correlation, correlation break, global equity asset allocation, financial crises, regime switching

JEL Classification: G01, G11

Suggested Citation

Solnik, Bruno and Watewai, Thaisiri, Does Extreme Correlation Matter in Global Equity Asset Allocation? (February 16, 2018). Available at SSRN: https://ssrn.com/abstract=3124858 or http://dx.doi.org/10.2139/ssrn.3124858

Bruno Solnik (Contact Author)

Hong Kong University of Science & Technology (HKUST) - Department of Finance ( email )

Clear Water Bay, Kowloon
Hong Kong

HEC Paris - Departement Finance et Economie ( email )

1, rue de la Liberation
Jouy-en-Josas Cedex, 78351
France
+33 1 39 67 72 84 (Phone)
+33 1 39 67 70 85 (Fax)

Thaisiri Watewai

Chulalongkorn University - Department of Banking & Finance ( email )

Thailand

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

Paper statistics

Downloads
203
Abstract Views
700
rank
191,095
PlumX Metrics