Deep learning with long short-term memory networks for financial market predictions
By Thomas G. Fischer, Christopher Krauss. 2017
2,474 citations recorded by the source corpus.
Long short-term memory (LSTM) networks are a state-of-the-art technique for sequence learning. They are less commonly applied to financial time series predictions, yet inherently suitable for this domain. We deploy LSTM networks for predicting out-of-sample directional movements for the constituent stocks of the S&P 500 from 1992 until 2015. With daily returns of 0.46 percent and a Sharpe ratio of 5.8 prior to transaction costs, we find LSTM networks to outperform memory-free classification methods, i.e., a random forest (RAF), a deep neural net (DNN), and a logistic regression classifier (LOG). The outperformance relative to the general market is very clear from 1992 to 2009, but as of 2010, excess returns seem to have been arbitraged away with LSTM profitability fluctuating around zero after transaction costs. We further unveil sources of profitability, thereby shedding light into the black box of artificial neural networks. Specifically, we find one common pattern among the stocks selected for trading – they exhibit high volatility and a short-term reversal return profile. Leveraging these findings, we are able to formalize a rules-based short-term reversal strategy that yields 0.23 percent prior to transaction costs. Further regression analysis unveils low exposure of the LSTM returns to common sources of systematic risk – also compared to the three benchmark models.
Subfields: Economics and Econometrics, Finance, Management Science and Operations Research, Model Architectures, Mathematical Optimization
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Economics and Econometrics
- Omer Berat Sezer
- A. Ozbayoglu
- Weiwei Jiang
- M. Ugur Gudelek
- Alexander Deinert
- Thomas G. Fischer
- Sima Siami‐Namini
- Akbar Siami Namin
In Finance
- Weiwei Jiang
- A. Ozbayoglu
- Omer Berat Sezer
- M. Ugur Gudelek
- Mathias Kraus
- Alexander Deinert
- Thomas G. Fischer
- Sima Siami‐Namini
In Management Science and Operations Research
- Omer Berat Sezer
- Weiwei Jiang
- A. Ozbayoglu
- M. Ugur Gudelek
- Sima Siami‐Namini
- Akbar Siami Namin
- Mathias Kraus
- Neda Tavakoli
In Model Architectures
- Weiwei Jiang
- A. Ozbayoglu
- Omer Berat Sezer
- M. Ugur Gudelek
- Sima Siami‐Namini
- Akbar Siami Namin
- Neda Tavakoli
- Cornelis W. Oosterlee
In Mathematical Optimization
- Weiwei Jiang
- A. Ozbayoglu
- Omer Berat Sezer
- Sima Siami‐Namini
- Akbar Siami Namin
- Cornelis W. Oosterlee
- Sander M. Bohte
- Neda Tavakoli
Citations in the corpus
- Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, methodology): “The authors of [123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 114] used S&P500 as their dataset.”
- Conditional Time Series Forecasting with Convolutional Neural Networks (2017, methodology): “Currently, recurrent neural networks (RNNs), and in particular the long-short term memory unit (LSTM) [12], [5], are the state-of-the-art in time series forecasting, see also [14] and in particular [7] for financial forecasting results.”
- Deep Learning for Financial Applications : A Survey (2020, neutral): “[41] S&P500 Index 1989-2005 Price data, Volume LSTM Return, STD, SR, Accuracy Python, TensorFlow, Keras, R, H2O”
- Financial Trading Model with Stock Bar Chart Image Time Series with Deep Convolutional Neural Networks (2019, methodology): “Fischer et al. [9] used LSTM to predict out-of-sample directional movements for stocks of the S&P500 from 1992 until 2015.”
- Statistical Arbitrage in Cryptocurrency Markets (2019, methodology): “As random forest implementation, we use Pedregosa et al. (2011) and largely follow Fischer and Krauss (2018) and Krauss et al. (2017) with respect to the parameters of random forest model.”
- Long short‐term memory networks for CSI300 volatility prediction with Baidu search volume (2018, background): “al(2017) applied LSTM networks to all S&P 500 constituents from 1992 until 2015 [4].”
- Applications of deep learning in stock market prediction: recent progress (2020, methodology): “…Zhang et al. (2019a); Zhou et al. (2019a) Maximum Drawdown Zhou et al. (2019a) Annualized Volatility Karathanasopoulos & Osman (2019) Sharpe Ratio Chen et al. (2017); Fischer & Krauss (2018); Sezer & Ozbayoglu (2018); Hoseinzade & Haratizadeh (2019); Karathanasopoulos & Osman (2019); Matsunaga…”
- A Deep Collocation Method for the Bending Analysis of Kirchhoff Plate (2021, background): “Several types of deep neural networks such as convolutional neural networks (CNN) and recurrent/recursive neural networks (RNN) [16] have been created, which further boost the application of deep learning in image processing [17], object detection [18], speech recognition [19] and many other domains”
- A Review of Deep Learning with Special Emphasis on Architectures, Applications and Recent Trends (2019, methodology): “Fisher and Krauss [157] adapted LSTM for stock prediction and compared its performance with memory-free based algorithms like random forest, logistic regression classifier and deep neural network.”
- Temporal Convolutional Networks Applied to Energy-Related Time Series Forecasting (2020, background): “It has been considered a standard forecasting model for several tasks such as traffic prediction [13], solar power forecasting [14], financial market predictions [15], and electricity price prediction [16].”
- DeepLOB: Deep Convolutional Neural Networks for Limit Order Books (2018, neutral): “Unlike CNNs which are less widely applied in financial markets, the LSTM has been popular in recent years, [42, 28, 43, 44, 45, 46, 47, 20] all utilising LSTMs to analyse financial data.”
- AccPar: Tensor Partitioning for Heterogeneous Deep Learning Accelerators (2020, background): “Advances in deep learning (DL) have become the main drivers of revolutions in various commercial and enterprise applications, such as computer vision [1–3], social network [4– 6], financial data analysis [7–9], healthcare [10–12] and sci-entific computing [13–15].”
- Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation (2018, background): “The RNNs can be trained to learn sequential or time-varying patterns by facilitate time delay units through feedback connections.”