Multi-Period Asset Allocation Under Hidden Markovianly Driven Noises

Abstract

Techniques of filtering and parameter reestimation of a general hidden Markov model are developed and applied to a discrete time multi-period asset allocation problem, where a commonly used mean-variance utility is considered and recursive calculation of an explicit optimal portfolio is provided. Our result is a generalization of that by Robert J. Elliott and John van der Hoek.

Keywords:

• Asset allocation
• EM algorithm
• Hidden Markov filter
• Hidden Markov model
• Mean-variance portfolio selection

Mathematics Subject Classification:

• 60J05
• 60G35

Acknowledgments

The author would like to thank Professor Allanus Tsoi (University of Missouri – Columbia) for the helpful discussions.