Forecasting oil & gas etfs´ price movements using convolutional neural networks

• Autor(a) principal: Serafin, Marc Lorenzo
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/156853
• topic: Forecasting; Deep learning; Technical analysis; Oil & gas industry; Convolutional neural networks; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

Thanks to advances in processing power, we have seen the revival of artificial intelligence after the 1980s, and algorithmic trading has become quite popular in the last two decades. In this paper, a convolutional neural network for image recognition was constructed. The CNN recognises patterns in 2D images generated from financial data and classifies them as BUY, SELL or HOLD. The analysed ETF, XLE, is from the Oil & Gas sector. The results are evaluated computationally and financially and compared to other industries. Overall, the CNN approach seems promising but generally, it was not possible to outperform the Buy&Hold strategy.