A model based on stochastic approximation is proposed to minimize the risk of portfolios of financial assets. Value-at-Risk, a widely used risk measure in financial management practices, has rarely been used in optimal portfolio context selection due to difficulties in computational implementation. The proposed model allows the minimum Value-at-Risk portfolio problem to be solved through an approximation of the objective function. This paper analyzes the performance of the model in a portfolio of Brazilian market assets.
portfolio management; finance; optimization