How much foreign stocks? : classical versus Bayesian approaches to asset allocation

  • The classical approaches to asset allocation give very different conclusions about how much foreign stocks a US investor should hold. US investors should either allocate a large portion of about 40% to foreign stocks (which is the result of mean/variance optimization and the international CAPM) or they should hold no foreign stocks at all (which is the conclusion of the domestic CAPM and mean/variance spanning tests). There is no way in between. The idea of the Bayesian approach discussed in this article is to shrink the mean/variance efficient portfolio towards the market portfolio. The shrinkage effect is determined by the investor's prior belief in the efficiency of the market portfolio and by the degree of violation of the CAPM in the sample. Interestingly, this Bayesian approach leads to the same implications for asset allocation as the mean-variance/tracking error criterion. In both cases, the optimal portfolio is a combination of the market portfolio and the mean/variance efficient portfolio with the highest Sharpe ratio. Applying both approaches to the subject of international diversification, we find that a substantial home bias is only justified when a US investor has a strong belief in the global mean/variance efficiency of the US market portfolio and when he has a high regret aversion of falling behind the US market portfolio. We also find that the current level of home bias can be justified whenever-regret aversion is significantly higher than risk aversion. Finally, we compare the Bayesian approach of shrinking the mean/variance efficient portfolio towards the market portfolio to another Bayesian approach which shrinks the mean/variance efficient portfolio towards the minimum-variance portfolio. An empirical out-of-sample study shows that both Bayesian approaches lead to a clearly superior performance compared to the classical mean/variance efficient portfolio.

Download full text files

Export metadata

Additional Services

Share in Twitter Search Google Scholar
Metadaten
Author:Ulf HeroldGND, Raimond MaurerORCiDGND
URN:urn:nbn:de:hebis:30:3-260106
ISSN:1434-3401
Parent Title (Portuguese):Goethe-Universität Frankfurt am Main. Fachbereich Wirtschaftswissenschaften: Working paper series / Finance & accounting ; No. 92
Series (Serial Number):Working paper series / Johann-Wolfgang-Goethe-Universität Frankfurt am Main, Fachbereich Wirtschaftswissenschaften : Finance & Accounting (92)
Publisher:Goethe-Universität Frankfurt, Fachbereich Wirtschaftswissenschaften
Place of publication:Frankfurt am Main
Document Type:Working Paper
Language:English
Year of Completion:2003
Year of first Publication:2003
Publishing Institution:Universitätsbibliothek Johann Christian Senckenberg
Release Date:2020/09/15
Tag:Bayesian inference; asset-pricing models; estimation risk; international diversification; portfolio selection
GND Keyword:Portfoliomanagement; Aktie; Capital-Asset-Pricing-Modell; Bayes-Regel
Page Number:36
HeBIS-PPN:470270527
Institutes:Wirtschaftswissenschaften / Wirtschaftswissenschaften
Dewey Decimal Classification:3 Sozialwissenschaften / 33 Wirtschaft / 330 Wirtschaft
Sammlungen:Universitätspublikationen
Licence (German):License LogoDeutsches Urheberrecht