A new approach to robust modeling of the multi- period portfolio problem

Abstract

In this paper, we developed a new robust model of multi-period portfolio problem. One of the key concerns in any asset allocation problem is how to cope with uncertainty about future returns. There are some approaches in the literature for this purpose including stochastic programming and robust optimization. Applying these techniques to multi-period portfolio problem may increase the problem size in a way that the resulting model is intractable. In this paper, we proposed a novel approach to formulate multi-period portfolio problem as an uncertain linear program assuming that asset return follows the single-index factor model. We also used robust optimization technique to solve the resulted problem. In order to evaluate the performance of the proposed model, we applied a numerical example using simulated data. Key words: Robust optimization, multi-period portfolio problem, asset pricing model, parameter uncertainty.