TY - JOUR T1 - A New Diagnostic Approach to Evaluating the Stability of Optimal Portfolios JF - The Journal of Investing SP - 37 LP - 45 DO - 10.3905/joi.2016.25.1.037 VL - 25 IS - 1 AU - Zhongjin Yang AU - Keli Han AU - Marat Molyboga AU - Georgiy Molyboga Y1 - 2016/02/29 UR - https://pm-research.com/content/25/1/37.abstract N2 - We introduce a new quantitative approach that can be used as a diagnostic tool for measuring the stability of optimal portfolio weights for a very general set of mean-variance optimization methods. We present a derivation of the approach within a numerical analysis framework and demonstrate this method’s benefits using a few common examples of shrinkage estimators of the correlation matrix and volatility vector. Our technique has practical importance in evaluating the improvements in stability gained by employing various statistical estimators of covariance matrixes without having to perform complex calculations or use numerical simulations.TOPICS: Statistical methods, portfolio construction ER -