Abstract
This article presents a new approach for building robust portfoliosbased on stochastic efficiency analysis, by using the ChanceConstrained Data Envelopment Analysis (CCDEA) model and peri-ods of market downturn, i.e. worst-state market. The model is ableto accommodate investors who exhibit different risk behaviors andthe empirical analysis is done on assets traded on the Brazil StockExchange, B3 (Brasil, Bolsa, Balc~ao).
The results confirm that theproposed model achieved portfolios that at the same time reducedsystematic risk and maximized portfolio returns when working withworse market state data and higher levels of risk aversion. A higherlevel of risk aversion also led to better risk-return ratios, which canbe seen in higher Sharpe ratio values.