Linking Individual Investors’ Preferences to a Portfolio Optimization Model

When optimizing a portfolio, individual investors today already knew potential returns from the capital are often offset by the amount of risk willing to take; hence, to lower the risk associated with the investment, they’ll have to diversify through a pool of portfolio from different asset classes (such as stocks, bonds, mutual funds, and cash, etc.). Since most studies have disregarded preferences of individual investors and selections of different asset classes in model formulations, this study provides a systematic approach to set priorities among multi-criteria and trade-off among objectives for Taiwanese individual investors. For that, the Analytic Network Process (ANP) is suggested to determine an asset allocation scheme tailored to the specific requirements of individual investors. Such scheme is then applied to the Markowitz model of portfolio optimization. The Variable Neighborhood Search (VNS) algorithm is constructed to build an efficient frontier of multiple portfolios which offer investors more alternatives on asset selections.