What question did this study set out to answer?

The aim is to assess the effectiveness of GRU and LSTM models in forecasting hourly returns of cryptocurrencies.

February 2, 20260 citations

Forecasting Hourly Cryptocurrency Returns Using Recurrent Neural Networks: A Comparative Study of GRU, LSTM and Classical Models

JMJakub MorkowskiPoznań University of Economics and Business

Key Points

The aim is to assess the effectiveness of GRU and LSTM models in forecasting hourly returns of cryptocurrencies.
Evaluated short-horizon forecasts using GRU and LSTM architectures.
Compared performance across multiple cryptocurrencies: Bitcoin, Ethereum, Binance Coin, Litecoin.
Utilized a unified evaluation protocol with identical data coverage and various error metrics (RMSE, MAE, DA).
Implemented classical benchmarks (ARIMA, GARCH) under the same forecasting design.
GRU networks consistently achieved lower error rates compared to LSTM and classical models.
Best GRU configurations showed directional accuracy (DA) between 0.65 and 0.72 depending on the asset.
Demonstrated the suitability of recurrent gating for the volatile nature of cryptocurrency returns.

Abstract

This study evaluates short-horizon forecasting of hourly cryptocurrency returns using two recurrent neural architectures—GRU and LSTM—estimated in more than 500 configurations across Bitcoin, Ethereum, Binance Coin and Litecoin. We adopt a unified protocol with intersection evaluation windows to ensure identical data coverage across models, and we compare magnitude-based errors (RMSE, MAE, MASE, sMAPE) with direction-based performance (Directional Accuracy, DA). Classical benchmarks (ARIMA/ETS, GARCH and a Random Walk random-walk) are estimated under the same one-step-ahead design. Empirically, GRU networks consistently achieve lower errors and higher DA than LSTM and traditional models. Best GRU configurations reach DA ≈ 0.65–0.72 depending on the asset, while requiring smaller amplitude recalibration. The results indicate that parsimonious recurrent gating is well-suited to the high-volatility, short-memory structure of cryptocurrency returns. Methodologically, the paper replicates and extends a previously published currency-market framework to a more turbulent domain, reinforcing the external validity of the findings.

问 AI

Bookmark

View Full Paper