Dramatically Improved Portfolio Optimization Results with Noise-Filtered Covariance : Paper 6 in a Series 'Uncovering Hidden Dependencies'

Alexander Izmailov, Ph.D (theoretical physics) and Brian Shay, Ph.D (mathematics) of Market Memory Trading, L.L.C., present in a series of nine (9) white papers, aspects of a revolutionary advance in uncovering hidden dependencies via filtering noise from correlation matrices developed by the New York based company, Market Memory Trading, L.L.C. (MMT). Correlations are quantitative measures of these dependencies and noise filtering increases their accuracy as a decision-making tool, from asset allocation to LIBOR Surveillance and cyber security.“DRAMATICALLY IMPROVED PORTFOLIO OPTIMIZATION RESULTS WITH NOISE-FILTERED COVARIANCE.” White Paper 6, dated January 20, 2013, provides a demonstration that in-sample Markowitz type mean-variance optimization, carried out with noise filtered covariance matrices, results in asset allocation that leads to 2-3 times increase of the Sharpe ratio compared to the same optimization carried out without noise filtering; and a demonstration of 2-3 times increase of the Sharpe ratio due to asset allocation obtained via optimization, carried out with noise filtered covariance matrices, for two possible optimization scenarios – maximization of portfolio return at a fixed volatility and minimization of portfolio volatility at a fixed return. Refer to Appendix A for Complete Series