Uso de redes neurais profundas para previsão de curto prazo do preço da criptomoeda ethereum

Abstract

Cryptocurrency has become a popular asset in global financial markets, meaning that not only individual investors but also asset management companies around the world are considering this new investment class. The main contribution of this research is to address an intra-day forecasting problem with hourly granularity by comparing deep network architectures, including ones with and attention mechanisms for the Ethereum intrinsic cryptocurrency (ETH). The results showed that the TCN outperformed other architectures considered for a short-term forecast period in terms of processing time and it is amongst the most accurate models using an ARIMA model as a baseline.