{"type":"ScholarlyArticle","title":"Multi-scale Two-way Deep Neural Network for Stock Trend Prediction","authors":[{"name":"Guang Liu","slug":"guang-liu","given":"Guang","family":"Liu","affiliation":"PingAn Life Insurance Company of China , Ltd.; School of Computer Science , Beijing University of Posts and Telecommunications"},{"name":"Yuzhao Mao","slug":"yuzhao-mao","given":"Yuzhao","family":"Mao","affiliation":"PingAn Life Insurance Company of China , Ltd.; School of Computer Science , Beijing University of Posts and Telecommunications"},{"name":"Qi Sun","slug":"qi-sun","given":"Qi","family":"Sun","affiliation":"PingAn Life Insurance Company of China , Ltd."},{"name":"Hailong Huang","slug":"hailong-huang","given":"Hailong","family":"Huang","affiliation":"PingAn Life Insurance Company of China , Ltd."},{"name":"Weiguo Gao","slug":"weiguo-gao","given":"Weiguo","family":"Gao","affiliation":"PingAn Life Insurance Company of China , Ltd."},{"name":"Xuan Li","slug":"xuan-li","given":"Xuan","family":"Li","affiliation":"PingAn Life Insurance Company of China , Ltd."},{"name":"Jianping Shen","slug":"jianping-shen","given":"Jianping","family":"Shen","affiliation":"PingAn Life Insurance Company of China , Ltd."},{"name":"Ruifan Li","slug":"ruifan-li","given":"Ruifan","family":"Li","affiliation":"School of Computer Science , Beijing University of Posts and Telecommunications"},{"name":"Xiaojie Wang","slug":"xiaojie-wang","given":"Xiaojie","family":"Wang","affiliation":"School of Computer Science , Beijing University of Posts and Telecommunications"}],"abstract":"Stock Trend Prediction(STP) has drawn wide attention from various fields, especially Artificial Intelligence. Most previous studies are single-scale oriented which results in information loss from a multi-scale perspective. In fact, multi-scale behavior is vital for making intelligent investment decisions. A mature investor will thoroughly investigate the state of a stock market at various time scales. To automatically learn the multi-scale information in stock data, we propose a Multi-scale Two-way Deep Neural Network. It learns multi-scale patterns from two types of scale-information, wavelet-based and downsampling-based, by eXtreme Gradient Boosting and Recurrent Convolutional Neural Network, respectively. After combining the learned patterns from the two-way, our model achieves state-of-the-art performance on FI-2010 and CSI-2016, where the latter is our published long-range stock dataset to help future studies for STP task. Extensive experimental results on the two datasets indicate that multi-scale information can significantly improve the STP performance and our model is superior in capturing such information.","doi":"10.24963/ijcai.2020/628","url":"https://ijcai.org/proceedings/2020/628","pdf":"https://ijcai.org/proceedings/2020/0628.pdf","markdown":"https://ijcai.org/proceedings/2020/628.md","bibtex":"https://ijcai.org/proceedings/2020/bibtex/628","volume":{"slug":"2020","title":"Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2020","url":"https://ijcai.org/proceedings/2020/","year":2020,"location":"Yokohama","dates":"January 2021 (postponed from 11-17 July 2020)","editors":"Christian Bessiere","isbn":"978-0-9992411-6-5","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2020"},"section":"Special Track on AI in FinTech","track":"Special Track on AI in FinTech","number":628,"pages":{"first":4555,"last":4561},"keywords":["Foundation for AI in FinTech: Data mining and knowledge discovery for FinTech","Foundation for AI in FinTech: Analyzing big financial data","AI for trading: AI for novel financial models","Foundation for AI in FinTech: General","AI for risk and security: AI for market movement and change analysis","Other areas: Financial decision-support system","Foundation for AI in FinTech: Modeling financial market microstructure"],"videos":[{"label":"Video","url":"https://videos.ijcai20.org/24561/file_24561.m3u8","length_sec":258,"provider":"hls"}],"references":[{"position":1,"raw":"[Chen and Guestrin, 2016] Tianqi Chen and Carlos Guestrin. Xgboost: A scalable tree boosting system. In SIGKDD 2016, pages 785-794. ACM, 2016.","title":"Xgboost: A scalable tree boosting system","authors":"Tianqi Chen, Carlos Guestrin","year":"2016","venue":"SIGKDD 2016","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Cui et al., 2016] Zhicheng Cui, Wenlin Chen, and Yixin Chen. Multi-scale convolutional neural networks for time series classification. arXiv preprint arXiv:1603.06995, 2016.","title":"Multi-scale convolutional neural networks for time series classification","authors":"Zhicheng Cui, Wenlin Chen, Yixin Chen","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Dacorogna et al., 1996] Michel M Dacorogna, Cindy L Gauvreau, Ulrich A Müller, Richard B Olsen, and Olivier V Pictet. Changing time scale for short-term forecasting in financial markets. Journal of Forecasting, 15(3):203-227, 1996.","title":"Changing time scale for short-term forecasting in financial markets","authors":"Cindy L Michel M Dacorogna, Ulrich A Gauvreau, Richard B Müller, Olivier V Olsen, Pictet","year":"1996","venue":"Journal of Forecasting","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Deng et al., 2017] Yue Deng, Zhiquan Ren, Youyong Kong, Feng Bao, and Qionghai Dai. A hierarchical fused fuzzy deep neural network for data classification. IEEE Trans- actions on Fuzzy Systems, 25(4):1006-1012, 2017.","title":"A hierarchical fused fuzzy deep neural network for data classification","authors":"Yue Deng, Zhiquan Ren, Youyong Kong, Feng Bao, Qionghai Dai","year":"2017","venue":"IEEE Transactions on Fuzzy Systems","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Di Persio and Honchar, 2016] Luca Di Persio and Olek- sandr Honchar. Artificial neural networks architectures for stock price prediction: Comparisons and applications. In- ternational journal of circuits, systems and signal process- ing, 10:403-413, 2016.","title":"Artificial neural networks architectures for stock price prediction: Comparisons and applications","authors":"Di Persio, ; Honchar, Di Luca, Oleksandr Persio, Honchar","year":"2016","venue":"International journal of circuits, systems and signal processing","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Ehsan and Saman, 2019] Ehsan and Saman. Cnnpred: Cnn- based stock market prediction using a diverse set of vari- ables. Expert Systems with Applications, 129:273-285, 2019.","title":"Cnnpred: Cnnbased stock market prediction using a diverse set of variables","authors":"Saman ; Ehsan, Saman Ehsan","year":"2019","venue":"Expert Systems with Applications","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Fernández et al., 2019] César Fernández, Luis Salinas, and Claudio E Torres. A meta extreme learning machine method for forecasting financial time series. Applied In- telligence, 49(2):532-554, 2019.","title":"A meta extreme learning machine method for forecasting financial time series","authors":"César Fernández, Luis Salinas, Claudio E Torres","year":"2019","venue":"Applied Intelligence","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Geva, 1998] Amir B Geva. Scalenet-multiscale neural- network architecture for time series prediction. IEEE Transactions on neural networks, 9(6):1471-1482, 1998.","title":"Scalenet-multiscale neuralnetwork architecture for time series prediction","authors":"B Amir, Geva","year":"1998","venue":"IEEE Transactions on neural networks","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Gunduz et al., 2017] Hakan Gunduz, Yusuf Yaslan, and Zehra Cataltepe. Intraday prediction of borsa istanbul us- ing convolutional neural networks and feature correlations. Knowledge-Based Systems, 137:138-148, 2017.","title":"Intraday prediction of borsa istanbul using convolutional neural networks and feature correlations","authors":"Hakan Gunduz, Yusuf Yaslan, Zehra Cataltepe","year":"2017","venue":"Knowledge-Based Systems","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Hu and Qi, 2017] Hao Hu and Guo-Jun Qi. State-frequency memory recurrent neural networks. In ICML, pages 1568- 1577, 2017.","title":"State-frequency memory recurrent neural networks","authors":"Hao Hu, Guo-Jun Qi","year":"2017","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Kara et al., 2011] Yakup Kara, Melek Acar Boyacioglu, and Ömer Kaan Baykan. Predicting direction of stock price index movement using artificial neural networks and sup- port vector machines: The sample of the istanbul stock exchange. Expert systems with Applications, 38(5):5311- 5319, 2011.","title":"Predicting direction of stock price index movement using artificial neural networks and support vector machines: The sample of the istanbul stock exchange","authors":"Yakup Kara, Melek Acar Boyacioglu, Ömer Kaan Baykan","year":"2011","venue":"Expert systems with Applications","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Kim, 2003] Kyoung-jae Kim. Financial time series fore- casting using support vector machines. Neurocomputing, 55(1-2):307-319, 2003.","title":"Financial time series forecasting using support vector machines","authors":"Kyoung-Jae Kim","year":"2003","venue":"Neurocomputing","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Lahmiri, 2014] Salim Lahmiri. Wavelet low-and high- frequency components as features for predicting stock prices with backpropagation neural networks. Journal of King Saud University-Computer and Information Sci- ences, 26(2):218-227, 2014.","title":"Wavelet low-and highfrequency components as features for predicting stock prices with backpropagation neural networks","authors":"Salim Lahmiri","year":"2014","venue":"Journal of King Saud University-Computer and Information Sciences","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Li et al., 2016] Xiaodong Li, Haoran Xie, Ran Wang, Yi Cai, Jingjing Cao, Feng Wang, Huaqing Min, and Xi- aotie Deng. Empirical analysis: stock market prediction via extreme learning machine. Neural Computing and Ap- plications, 27(1):67-78, 2016.","title":"Empirical analysis: stock market prediction via extreme learning machine","authors":"Xiaodong Li, Haoran Xie, Ran Wang, Yi Cai, Jingjing Cao, Feng Wang, Huaqing Min, Xiaotie Deng","year":"2016","venue":"Neural Computing and Applications","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Lin et al., 2017] Tao Lin, Tian Guo, and Karl Aberer. Hy- brid neural networks for learning the trend in time series. In IJCAI-17, pages 2273-2279, 2017.","title":"Hybrid neural networks for learning the trend in time series","authors":"Tao Lin, Tian Guo, Karl Aberer","year":"2017","venue":"IJCAI-17","doi":null,"url":null,"internal_url":"/proceedings/2017/316"},{"position":16,"raw":"[Luo and Yu, 2019] Wei Luo and Feng Yu. Recurrent high- way networks with grouped auxiliary memory. IEEE Ac- cess, 7:182037-182049, 2019.","title":"Recurrent highway networks with grouped auxiliary memory","authors":"Wei Luo, Feng Yu","year":"2019","venue":"IEEE Access","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Ntakaris et al., 2018] Adamantios Ntakaris, Martin Magris, Juho Kanniainen, Moncef Gabbouj, and Alexandros Iosi- fidis. Benchmark dataset for mid-price forecasting of limit order book data with machine learning methods. Journal of Forecasting, 37(8):852-866, 2018.","title":"Benchmark dataset for mid-price forecasting of limit order book data with machine learning methods","authors":"Ntakaris","year":"2018","venue":"Journal of Forecasting","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Papadimitriou and Yu, 2006] Spiros Papadimitriou and Philip Yu. Optimal multi-scale patterns in time series streams. In SIGMOD 2006, pages 647-658. ACM, 2006.","title":"Optimal multi-scale patterns in time series streams","authors":"Spiros Papadimitriou, Philip Yu","year":"2006","venue":"SIGMOD 2006","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Patel et al., 2015] Jigar Patel, Sahil Shah, Priyank Thakkar, and K Kotecha. Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques. Expert Systems with Appli- cations, 42(1):259-268, 2015.","title":"Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques","authors":"Jigar Patel, Sahil Shah, Priyank Thakkar, Kotecha","year":"2015","venue":"Expert Systems with Applications","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Tran et al., 2018] Dat Thanh Tran, Alexandros Iosifidis, Juho Kanniainen, and Moncef Gabbouj. Temporal attention-augmented bilinear network for financial time- series data analysis. IEEE transactions on neural networks and learning systems, 30(5):1407-1418, 2018.","title":"Temporal attention-augmented bilinear network for financial timeseries data analysis","authors":"Tran","year":"2018","venue":"IEEE transactions on neural networks and learning systems","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Tsai and Hsiao, 2010] Chih-Fong Tsai and Yu-Chieh Hsiao. Combining multiple feature selection methods for stock prediction: Union, intersection, and multi-intersection approaches. Decision Support Systems, 50(1):258-269, 2010.","title":"Combining multiple feature selection methods for stock prediction: Union, intersection, and multi-intersection approaches","authors":"Hsiao ; ] Chih-Fong Tsai, Yu-Chieh Tsai, Hsiao","year":"2010","venue":"Decision Support Systems","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Tsantekidis et al., 2017a] Avraam Tsantekidis, Nikolaos Passalis, Anastasios Tefas, Juho Kanniainen, Moncef Gab- bouj, and Alexandros Iosifidis. Forecasting stock prices from the limit order book using convolutional neural net- works. In CBI 2017, volume 1, pages 7-12. IEEE, 2017.","title":"Forecasting stock prices from the limit order book using convolutional neural networks","authors":"Tsantekidis","year":"2017","venue":"CBI 2017","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Tsantekidis et al., 2017b] Avraam Tsantekidis, Nikolaos Passalis, Anastasios Tefas, Juho Kanniainen, Moncef Gab- bouj, and Alexandros Iosifidis. Using deep learning to de- tect price change indications in financial markets. In EU- SIPCO 2017, pages 2511-2515. IEEE, 2017.","title":"Using deep learning to detect price change indications in financial markets","authors":"Tsantekidis","year":"2017","venue":"EU-SIPCO 2017","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Tsantekidis et al., 2018] Avraam Tsantekidis, Nikolaos Passalis, Anastasios Tefas, Juho Kanniainen, Moncef Gabbouj, and Alexandros Iosifidis. Using deep learning for price prediction by exploiting stationary limit order book features. arXiv preprint arXiv:1810.09965, 2018.","title":"Using deep learning for price prediction by exploiting stationary limit order book features","authors":"Avraam Tsantekidis, Nikolaos Passalis, Anastasios Tefas, Juho Kanniainen, Moncef Gabbouj, Alexandros Iosifidis","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zhang et al., 2019] Zihao Zhang, Stefan Zohren, and Stephen Roberts. Deeplob: Deep convolutional neural net- works for limit order books. IEEE Transactions on Signal Processing, 67(11):3001-3012, 2019.","title":"Deeplob: Deep convolutional neural networks for limit order books","authors":"Zhang","year":"2019","venue":"IEEE Transactions on Signal Processing","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}