Bernardo Freitas Paulo da Costa
Professor of Mathematics at UFRJ, Brazil
Risk budgeting is a portfolio strategy where each asset contributes a pre-specified amount to the total portfolio risk. We propose a numerical framework in JuMP that uses only simulations of returns for estimating risk budgeting portfolios, and provide a Sample Average Approximation algorithm. We leveraged automatic differentiation and JuMP's modeling flexibility to build a clear and concise code. We also report on memory issues encountered when solving for every day in a 14 year horizon.