Loading...
Publication
Avaliação de Indicadores e Estratégias no Mercado Forex com base em Redes Neuronais
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 2.07 MB | Adobe PDF |
Authors
Abstract(s)
Esta dissertação analisa a eficácia de diferentes estratégias de análise técnica no mercado
Forex, utilizando Redes Neuronais para prever movimentos de preços. Foram testadas várias
estratégias de Médias Móveis Exponenciais, Relative Strength Index (RSI), Moving Average
Convergence / Divergence (MACD) e Bollinger Bands® nos pares cambiais EUR/USD,
EUR/ZAR, AUD/CAD, GBP/JPY, NZD/CHF, USD/BRL no período de 1999 a 2023, e
consideradas várias configurações de Redes Neuronais.
Os resultados indicam que as Redes Neuronais podem reproduzir relativamente bem a
estratégia ótima a adotar nos pares cambiais analisados. Foram obtidas taxas de precisão
média de 82%. A pesquisa contribui para a literatura sobre previsões financeiras e
negociação algorítmica, demonstrando o potencial das Redes Neuronais como ferramenta
para negociação no Forex. A dissertação conclui que, apesar das limitações atuais, as Redes
Neuronais têm potencial significativo para aplicação em estratégias de negociação no Forex,
mas requerem aprimoramento contínuo.
This dissertation analyzes the effectiveness of different technical analysis strategies in the Forex market, using neural networks to predict price movements. Various strategies of Exponential Moving Averages, Relative Strength Index (RSI), Moving Average Convergence / Divergence (MACD) and Bollinger Bands® were tested on the, EUR/USD, EUR/ZAR, AUD/CAD, GBP/JPY, NZD/CHF, USD/BRL currency pairs from 1999 to 2023, and various Neural Network configurations were considered. The results indicate that neural networks can reproduce relatively well the optimal strategy to adopt in the currency pairs analyzed. Average accuracy rates of 82% were obtained. The research contributes to the literature on financial forecasting and algorithmic trading, demonstrating the potential of neural networks as a tool for Forex trading. The dissertation concludes that, despite current limitations, neural networks have significant potential for application in Forex trading strategies but require continuous improvement.
This dissertation analyzes the effectiveness of different technical analysis strategies in the Forex market, using neural networks to predict price movements. Various strategies of Exponential Moving Averages, Relative Strength Index (RSI), Moving Average Convergence / Divergence (MACD) and Bollinger Bands® were tested on the, EUR/USD, EUR/ZAR, AUD/CAD, GBP/JPY, NZD/CHF, USD/BRL currency pairs from 1999 to 2023, and various Neural Network configurations were considered. The results indicate that neural networks can reproduce relatively well the optimal strategy to adopt in the currency pairs analyzed. Average accuracy rates of 82% were obtained. The research contributes to the literature on financial forecasting and algorithmic trading, demonstrating the potential of neural networks as a tool for Forex trading. The dissertation concludes that, despite current limitations, neural networks have significant potential for application in Forex trading strategies but require continuous improvement.
Description
Keywords
Forex Redes neuronais Análise técnica RSI MACD Bollinger Bands®