Systematic Review of Portfolio Optimization in the Brazilian Financial Market: Integrating News and Fundamental Indicators with Deep Reinforcement Learning
摘要
This study presents a systematic review of the literature on the application of Deep Reinforcement Learning (DRL) combined with news in Portuguese and fundamental quantitative analysis for optimizing financial asset portfolios. The review follows the Kitchenham model and was conducted using the Parsifal tool. Articles were collected from the IEEE Xplore, Scopus, ACM Digital Library, and Google Scholar databases. The findings indicate a growing interest in the use of DRL for portfolio optimization, with studies applying this technique alongside quantitative analysis reporting favorable results. However, no studies were identified that specifically address the integration of DRL with Portuguese news and fundamental indicators for portfolio optimization within the Brazilian financial market.