Strategic Asset Allocation for Long-Term Investors: Parameter Uncertainty and Prior Information

Abstract

We study the effect of parameter uncertainty on the long-run risk for three asset classes: stocks, bills and bonds. We model return dynamics as a Bayesian vector autoregression. With an uninformative prior we find that parameter uncertainty raises the annualised long-run volatilities of all three asset classes proportionally with the same factor relative to estimates that are conditional on ML parameter estimates. Correlations among returns appear robust against parameter uncertainty. As a result the effect of the investment horizon on optimal asset allocations is much weaker compared to models in which only equity returns are subject to parameter uncertainty. While equity is more volatile, the same holds for the other asset classes. Results are sensitive to alternative informative priors, but generally the term structure of risk for stocks and bonds are relatively flat for investment horizons up to 15 years.

Keywords: strategic asset allocation, bayesian vector autoregression, parameter uncertainty

JEL Classification: C32, G11, C11