Expert Systems With Applications
By Eunsuk Chong, Chulwoo Han, Frank C. Park
702 citations recorded by the source corpus.
Deep learning networks are applied to stock market analysis and prediction.A comprehensive analysis with different data representation methods is offered.Five-minute intraday data from the Korean KOSPI stock market is used.The network applied to residuals of autoregressive model improves prediction.Covariance estimation for market structure analysis is improved with the network. We offer a systematic analysis of the use of deep learning networks for stock market analysis and prediction. Its ability to extract features from a large set of raw data without relying on prior knowledge of predictors makes deep learning potentially attractive for stock market prediction at high frequencies. Deep learning algorithms vary considerably in the choice of network structure, activation function, and other model parameters, and their performance is known to depend heavily on the method of data representation. Our study attempts to provides a comprehensive and objective assessment of both the advantages and drawbacks of deep learning algorithms for stock market analysis and prediction. Using high-frequency intraday stock returns as input data, we examine the effects of three unsupervised feature extraction methodsprincipal component analysis, autoencoder, and the restricted Boltzmann machineon the networks overall ability to predict future market behavior. Empirical results suggest that deep neural networks can extract additional information from the residuals of the autoregressive model and improve prediction performance; the same cannot be said when the autoregressive model is applied to the residuals of the network. Covariance estimation is also noticeably improved when the predictive network is applied to covariance-based market structure analysis. Our study offers practical insights and potentially useful directions for further investigation into how deep learning networks can be effectively used for stock market analysis and prediction.
Subfields: Finance, Management Science and Operations Research, Knowledge Graphs & Representation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Finance
- Weiwei Jiang
- M. Ugur Gudelek
- A. Ozbayoglu
- Omer Berat Sezer
- Longbing Cao
- Ze Hu
- Matloob Khushi
- Yiqi Zhao
In Management Science and Operations Research
- Weiwei Jiang
- M. Ugur Gudelek
- Omer Berat Sezer
- A. Ozbayoglu
- Longbing Cao
- Ze Hu
- Matloob Khushi
- Yiqi Zhao
In Knowledge Graphs & Representation
- Dang Lien Minh
- Weiwei Jiang
- Ze Hu
- A. Ozbayoglu
- Longbing Cao
- Omer Berat Sezer
- Saman Haratizadeh
- M. Ugur Gudelek
Citations in the corpus
- Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, methodology): “In [80], DNN and lagged stock returns were used to predict the stock prices in The Korea Composite Stock Price Index (KOSPI).”
- Deep Learning Approach for Short-Term Stock Trends Prediction Based on Two-Stream Gated Recurrent Unit Network (2018, methodology): “On the other hand, the fundamental analysis [2], [5], [6] used natural language processing (NLP) to analyze financial news and financial statements from the company and predict the stock trends in the future (uptrend, down-trend).”
- Which Artificial Intelligence Algorithm Better Predicts the Chinese Stock Market? (2018, positive): “We find that DL is more effective when applied to high frequency data, consistent with the assumption of Chong et al. [22].”
- Stock Market Prediction on High-Frequency Data Using Generative Adversarial Nets (2018, methodology): “However, to our knowledge, most of these methods require expertise to impose specific restrictions on the input variables, such as combining related stocks together as entry data [12], inputting different index data to different layers of the deep neural network [34], and converting news text into s”
- CNNpred: CNN-based stock market prediction using a diverse set of variables (2019, methodology): “In (Chong et al., 2017), authors draw an analogy between different data representation methods including RBM, Auto-encoder and PCA applied on raw160 data with 380 variables.”
- Explainable Deep Learning: A Field Guide for the Uninitiated (2020, background): “…our pockets on our cell phones (Georgiev et al., 2017), in cars to help avoid car accidents (Jain et al., 2015), in banks to manage our investments (Chong et al., 2017) and evaluate loans (Pham & Shen, 2017), in hospitals to help doctors diagnose disease symptoms (Nie et al., 2015), at law…”
- Quantum computing for finance: Overview and prospects (2018, background): “These properties have made neural networks essential for, e.g., financial time-series prediction [65, 66].”
- Bankruptcy prediction using imaged financial ratios and convolutional neural networks (2019, neutral): “…on the prediction of stock price fluctuations (Yoshihara, Fujikawa, Seki, and Uehara, 2014; Aggarwal and Aggarwal, 2016; Persio and Honchar, 2017; Chong, Han, and Park, 2017; Bao, Yue, and Rao, 2017), where recurrent neural networks are often utilized for time series analysis.60 An example of…”
- Applications of deep learning in stock market prediction: recent progress (2020, methodology): “…et al. (2019); Ding & Qin (2019); Zhang et al. (2019b); Borovkova & Tsiamas (2019) Python PyTorch Yu & Wu (2019) Python Theano Zhang et al. (2017b); Chong et al. (2017); Siami- Namini et al. (2018); Pang et al. (2018); Tsantekidis et al. (2017a) Python scikit-learn Liu et al. (2019c); Eapen et…”
- A Survey of Forex and Stock Price Prediction Using Deep Learning (2021, methodology): “In this table, papers [22,24] used a CNN-based model; papers [32,33,35,38– 43,59] used an LSTM-based model; and papers [70,72] used a DNN-based model.”
- AI in Finance: Challenges, Techniques, and Opportunities (2021, methodology): “Typical embedding and representation techniques include those for word and text representations, neural language models, time series regression, sequential representation, and network and graph-based representations by classic network and graph theories, NN, and neural graphical networks [16, 44, 51”