Forex exchange rate forecasting using deep recurrent neural networks
- PDF / 1,567,153 Bytes
- 28 Pages / 439.37 x 666.142 pts Page_size
- 62 Downloads / 199 Views
Forex exchange rate forecasting using deep recurrent neural networks Alexander Jakob Dautel1 · Wolfgang Karl Härdle1,2 · Stefan Lessmann1 · Hsin‑Vonn Seow3 Received: 8 March 2019 / Accepted: 12 March 2020 © The Author(s) 2020
Abstract Deep learning has substantially advanced the state of the art in computer vision, natural language processing, and other fields. The paper examines the potential of deep learning for exchange rate forecasting. We systematically compare long shortterm memory networks and gated recurrent units to traditional recurrent network architectures as well as feedforward networks in terms of their directional forecasting accuracy and the profitability of trading model predictions. Empirical results indicate the suitability of deep networks for exchange rate forecasting in general but also evidence the difficulty of implementing and tuning corresponding architectures. Especially with regard to trading profit, a simpler neural network may perform as well as if not better than a more complex deep neural network. Keywords Deep learning · Financial time series forecasting · Recurrent neural networks · Foreign exchange rates JEL Classification C14 · C22 · C45
* Stefan Lessmann stefan.lessmann@hu‑berlin.de Alexander Jakob Dautel [email protected] Wolfgang Karl Härdle haerdle@hu‑berlin.de Hsin‑Vonn Seow Hsin‑[email protected] 1
School of Business and Economics, Humboldt-Universiät zu Berlin, Unter‑den‑Linden 6, 10099 Berlin, Germany
2
Singapore Management University, 50 Stamford Road, Singapore 178899, Singapore
3
Nottingham University Business School, 43500 Semenyih, Selangor Darul Ehsan, Malaysia
13
Vol.:(0123456789)
Digital Finance
1 Introduction Deep learning has revitalized research into artificial neural networks. Substantial methodological advancements associated with the optimization and regularization of large neural networks, the availability of large data sets together with the computational power to train large networks, and development of powerful, easyto-use software libraries, deep neural networks (DNNs) have achieved breakthrough performance in computer vision, natural language processing, and other domains (LeCun et al. 2015). A feature that sets deep learning apart from conventional machine learning is the ability automatically extract discriminative features from raw data (Nielsen 2015). Reducing the need for manual feature engineering, this ability decreases the costs of applying a learning algorithm in industry, simplifies tasks associated with model maintenance, and, more generally, broadens the scope of deep learning applications. Convolutional neural networks and recurrent neural networks (RNNs) have been particularly successful. The former represent the model of choice for computer vision tasks. RNNs are designed for processing sequential data including natural language, audio, and generally, any type of time series. The paper focuses on RNNs and examines their potential for financial time series forecasting. Deep-learning-based forecasting models
Data Loading...
