Elsevier

Applied Soft Computing

Financial time series forecasting with deep learning : A systematic literature review: 2005–2019

Highlights

We reviewed all searchable articles of deep learning (DL) for financial time series forecasting.

RNN based DL models (LSTM and GRU included) are the most common.

We compared DL models according to their performances in different forecasted asset classes.

To best of our knowledge, this is the first comprehensive DL survey for financial time series forecasting.

We provided current status of DL in financial time series forecasting, also highlighted the future opportunities.

Abstract

Financial time series forecasting is undoubtedly the top choice of computational intelligence for finance researchers in both academia and the finance industry due to its broad implementation areas and substantial impact. Machine Learning (ML) researchers have created various models, and a vast number of studies have been published accordingly. As such, a significant number of surveys exist covering ML studies on financial time series forecasting. Lately, Deep Learning (DL) models have appeared within the field, with results that significantly outperform their traditional ML counterparts. Even though there is a growing interest in developing models for financial time series forecasting, there is a lack of review papers that solely focus on DL for finance. Hence, the motivation of this paper is to provide a comprehensive literature review of DL studies on financial time series forecasting implementation. We not only categorized the studies according to their intended forecasting implementation areas, such as index, forex, and commodity forecasting, but we also grouped them based on their DL model choices, such as Convolutional Neural Networks (CNNs), Deep Belief Networks (DBNs), and Long-Short Term Memory (LSTM). We also tried to envision the future of the field by highlighting its possible setbacks and opportunities for the benefit of interested researchers.

Keywords

Deep learning

Finance

Computational intelligence

Machine learning

Time series forecasting

CNN

LSTM

RNN

Omer Berat Sezer received the B.S. degree in Electrical and Electronics Engineering (EEE) from Middle East Technical University (METU) NCC, Ankara, in 2009; M.S. degree in EEE from METU, Ankara, in 2013 with an emphasis on computer networks. He is graduated from Computer Engineering from TOBB ETU with Ph.D. degree, Ankara, in 2018, with an emphasis on machine learning and deep learning. He has worked as a software engineer and researcher at TUBITAK Space Technologies Research Institute, in Ankara for ten years. He is also now working as a software engineer in the Automotive Sector in Germany. His research interests are machine learning, artificial intelligence, time-series data analytics, computational intelligence, machine vision and software engineering.

Mehmet Ugur Gudelek received B.S. degree in Electrical Engineering from Middle East Technical University (METU), Ankara, Turkey, in 2015 with an emphasis on computers and M.S. degree in Computer Engineering from TOBB University of Economics and Technology (TOBB ETU), Ankara, Turkey, in 2019. He is currently a Computer Engineering Ph.D. candidate at TOBB ETU. He is also a teaching and research assistant at TOBB ETU. His research interests are machine learning, deep learning and time series data analytics.

Ahmet Murat Ozbayoglu graduated from the Department of Electrical Engineering at METU, Ankara, Turkey in 1991. He got his M.S. and Ph.D degrees from Systems Engineering at Missouri University of Science and Technology in 1993 and 1996, respectively. Then, he worked at Beyond Inc, St. Peters, MO as a product development engineer and consultant. In 2001 he joined MEMC Electronic Materials Inc., as a software engineer. In 2005, he joined the Department of Computer Engineering of TOBB University of Economics and Technology, Ankara. His research interests include machine learning, pattern recognition, big data, algorithmic trading, computational intelligence, machine vision.

View full text

© 2020 Elsevier B.V. All rights reserved.