Avaliação de estratégias de trading por algoritmo baseadas em rompimentos de linhas de tendências aplicadas ao índice de força relativa

Resumo

This dissertation evaluates the effectiveness of algorithmic trading strategies based on the breakout of trendlines applied to the Relative Strength Index (RSI). By leveraging historical data from stocks within the S&P 500 and adopting a rigorous approach to parameter optimization, the study provides a systematic analysis of how these signals can be used to generate buy and sell recommendations. The methodology, developed in Python, involves identifying trendlines derived from RSI, validating breakout signals, and backtesting their performance under various scenarios, including both in-sample and out-of-sample data. The results demonstrate that the proposed strategy consistently delivers superior returns compared to traditional buy-and-hold approaches in a significant number of cases. The Sharpe Ratio and annualized returns are particularly notable, showcasing the strategy’s strong risk-adjusted performance. Additionally, the flexibility of the model allows its extension to other asset classes, including indices, currencies, and intraday data, indicating the potential for broader applicability. The techniques developed for constructing trendlines provide a robust framework for future studies and can be valuable for other research efforts that incorporate trendlines as part of their signal generation processes, as well as, explores the application of trendline signals not directly to price charts but to derivative indicators such as RSI, showcasing the potential for generating robust trading signals from transformed data such as Moving Average Convergence Divergence (MACD), Stochastic Oscillator, Average True Range (ATR), and Bollinger Bands. A key aspect of this research is the practical implementation of alternating and randomized test intervals, ensuring randomness in their selection, the study reduces the risks associated with overfitting and enhances the robustness of the strategy’s evaluation. The results contribute to the field of technical analysis by implementing routines and techniques for the use of trendline breakouts applicable to many chart indicators. This study also lays the foundation for other research, including the integration of machine learning techniques and the application of these strategies in real-time trading environments.

Ficha do documento

Tipo

Dissertação

Ano

2025

Instituição

Fundação Getulio Vargas

Fonte

Repositório da FGV

Idioma

Português

Acesso

Acesso aberto

Identificador

oai:repositorio.fgv.br:10438/36558

Temas

EconomiaDados

Palavras-chave

AlgoritmoAnálise técnicaLinhas de tendênciaInvestimentosAçõesRSIÍndice de Força RelativaMercado financeiroPythonTradingBacktestingBacktestAlgorithmTechnical analysisTrendlinesInvestmentsStocksRelative strength indexEconomiaMercado financeiro - Inovações tecnológicasAlgoritmosInvestimentos - AnáliseInvestimentos - AdministraçãoÍndices de mercado de ações