Dynamic asset allocation strategy using a state-dependent Markov model: Applications to international equity markets

• Published in: Journal of international money and finance 2022-11, Vol.128, p.102705, Article 102705
• Main Authors: Hematizadeh, Roksana, Tajaddini, Reza, Hallahan, Terrence
• Format: Article
• Language: English
• Subjects: Dynamic asset allocation; Financial crisis; International markets; State-dependent strategy; Transaction costs
• ISSN: 0261-5606
• DOI: 10.1016/j.jimonfin.2022.102705

•We introduce a state-dependent Markov model strategy with the U.S. term spread as a phase identifier.•The U.S. monetary policies spill over into foreign markets, especially after the GFC.•The strategy remains profitable after accounting for transaction costs.•The strategy is able to set up selective portfolios in a bear market. This paper investigates the performance of an asset allocation strategy that relies on the state-dependent correlation between emerging and developed markets. We utilize a state-dependent Markov Model that employs a time-varying transition probability and uses the U.S. term spread differential as a market phase identifier. We show that by relying on the spillover effect of U.S. monetary policies on international equity markets, international investors can optimize returns on their investments by using a state-dependent model when diversifying their portfolios towards less volatile markets. Two states and optimal tangency portfolios reveal superior performance even after consideration of transaction costs. The effect of the U.S. term spread on equity returns is more apparent after the Global Financial Crisis; when positive (negative) term spread differentials are associated with an increased likelihood of being in a bull (bear) market.