Efficient Portfolios Computed via Moment-Based Bounding-approximations: Part I - EB

Abstract

We develop and analyze mean-variance efficient portfolios. Each portfolio comes as a solution of an optimization problem, which approximates the expected value of a utility function. The approximation is an upper bound on the expected value of the utility function. The bound is based on the first two probability moments and cross-moments of the portfolio”s random return. We prove that the optimal solution of the approximate optimization problem yields a mean-variance efficient portfolio. We illustrate how to use the resulting portfolio in practice by designing a daily trading strategy with stocks traded on the New York Stock Exchange (NYSE). The approximate optimization model is solved once every day. Out-of-sample numerical results are presented for 27 years of daily trading for 24 stocks from NYSE.