Asymmetric Risk and International Portfolio Choice

UTS Quantitative Finance Research Paper No. 160

31 Pages Posted: 26 May 2006

See all articles by Susan Thorp

Susan Thorp

The University of Sydney Business School

George Milunovich

Macquarie University - Department of Actuarial Studies and Business Analytics; Macquarie University, Macquarie Business School

Date Written: June 2005

Abstract

Empirical research shows that stock volatilities and correlations between markets rise more after negative shocks than after positive returns shocks of the same size. We measure the importance of these asymmetric effects for mean-variance investors holding portfolios of international equities who use dynamic conditional covariance forecasts to reweight their portfolios. Portfolio weights are computed using ex ante predictions from symmetric GARCH DCC and asymmetric GJR ADCC models, and a spectrum of expected returns. Data are weekly returns to equity price indices for the USA, Japan, UK and Australia. We find that the majority of realised portfolio standard deviations are less when we reweight using the asymmetric covariance model. Reductions in portfolio risk are significant according to Diebold-Mariano tests. Investors who are moderately risk averse and have longer rebalancing horizons benefit more from the asymmetric model than less risk averse, shorter-horizon investors, and would be prepared to pay up to 107 basis points annually to use it instead of the symmetric model. Benefits are greater for investors holding US equities.

Suggested Citation

Thorp, Susan and Milunovich, George, Asymmetric Risk and International Portfolio Choice (June 2005). UTS Quantitative Finance Research Paper No. 160, Available at SSRN: https://ssrn.com/abstract=904533 or http://dx.doi.org/10.2139/ssrn.904533

Susan Thorp (Contact Author)

The University of Sydney Business School ( email )

Abercrombie Building
H70
The University Of Sydney, NSW 2006
Australia
0290366354 (Phone)

George Milunovich

Macquarie University - Department of Actuarial Studies and Business Analytics ( email )

Australia

Macquarie University, Macquarie Business School ( email )

New South Wales 2109
Australia