Combining equilibrium, resampling, and analyst’s views in portfolio optimization

► We model a portfolio optimization methodology with equilibrium, and views and resampling. ► Methodology produces stable and diversified portfolios, and is theoretically grounded. ► Methodology is tested in a comprehensive sample of fixed income and equity indices. ► Methodology outperformed traditional approaches in 3 performance dimensions. This paper proposes the use of a portfolio optimization methodology which combines features of equilibrium models and investor’s views as in Black and Litterman (1992), and also deals with estimation risk as in Michaud (1998). In this way, our combined methodology is able to meet the needs of practitioners for stable and diversified portfolio allocations, while it is theoretically grounded on an equilibrium framework. We empirically test the methodology using a comprehensive sample of developed countries fixed income and equity indices, as well as sub-samples stratified by geographical region, time period, asset class and risk level. In general, our proposed combined methodology generates very competitive portfolios when compared to other methodologies, considering three evaluation dimensions: financial efficiency, diversification, and allocation stability. By generating financially efficient, stable, and diversified portfolio allocations, our methodology is suitable for long-term investors such as Central Banks and Sovereign Wealth Funds.