Dynamic Asset Allocation under Regime Switching Predictability and Parameter Uncertainty

Abstract

This paper solves the dynamic asset allocation problem under stock return predictability based on the dividend price ratio with regime shifts and parameter uncertainty in a fully Bayesian framework. Intertemporal hedging demands are simultaneously induced by predictability, regime shifts, parameter uncertainty, and learning about the regimes. Optimal policies display non-monotonic horizon effects whereby regime shifts tend to induce negative hedge demands in the short-run, while predictability induces positive hedge demands in the long-run. The economic costs of ignoring regime switching and predictability are high even in the light of regime and parameter uncertainty.