An asset allocation model with inequalities constraints and coherent risk measure: an application to Brazilian equities

Abstract

We propose a method for optimal portfolio selection built on the Black and Litterman model and with two major contributions. We introduce in the investors'objective function a risk measure named expected tail loss, which is useful in portfolio selection context as it supports the benefits of diversification and we allow investors'views to be expressed in terms of linear inequalities among expected returns, which seems more natural in the practice of portfolio selection. Further we implement the models using market database applied to Brazilian equities. The results show that our approach leads to lower risk optimal portfolios and that our proposed methodology to implement the investors'subjective views led to optimal portfolios with superior outcomes.