Stratetic Asset Allocation with an Arbitrage-Free Bond Market using Dynamic Programming

Recently, Campbell and Viceira (2002) have introduced a framework that allows for dynamic decisions in asset allocation. This paper follows up their work by showing how uncertainties and expectations may affect consumption and portfolio decisions in an intertemporal dynamic framework. We use the framework laid out by Merton(1973) for this type of model. We integrate into Merton's framework multi-factor models of bond pricing that satisfy the No-Arbitrage Principle. However, in this situation the intertemporal dynamic consumption and portfolio decision cannot be studied analytically by the integration of the multi-factors bond models as in Merton's original approach. In this paper the intertemporal optimization problem will be solved numerically using the method of the Dynamic Programming will be employed. The numerical errors due to the discretization of time and state space will be examined. Various properties of the optimal solution will be demonstrated numerically.