RGBSticks : A New Deep Learning Based Framework for Stock Market Analysis and Prediction

Abstract

We present a novel intuitive graphical representation for daily stock prices, which we refer as RGBSticks, a variation of classical candle sticks. This representation allows the usage of complex deep learning based techniques, such as deep convolutional autoencoders and deep convolutional generative adversarial networks to produce insightful visualizations for market's past and future states. We believe RGBStick representation has great potential to integrate human decision process and deep learning for stock market analysis and forecasting. The traders who are highly familiar with candlesticks are able to evaluate the results generated by deep learning algorithms by inspecting the varying color shades in a compact, instinctual and rapid fashion

Keywords

• artificial intelligence
• stock market
• time series

References

1. Morris G L. Candlestick Charting Explained: Timeless Techniques for Trading Stocks and Futures: Timeless Techniques for Trading stocks and Sutures; McGraw Hill Professional 2016.
2. Nison, S. Japanese candlestick charting techniques: a contemporary guide to the ancient investment techniques of the Far East; Penguin 2001.
3. Lewicki P, Czyzewska M, Hoffman H. Journal of Experimental Psychology Learning, Memory, and Cognition.; 13: 9 523 x. American Psychological Association 1987.
4. Berti A, Rizzolatti G. Visual processing without awareness: Evidence from unilateral neglect. Journal of cognitive neuroscience 1992; 4 (4) : 345-351
5. Chong E, Han C, Park F C. Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies. Expert Systems with Applications 2017; 83 : 187-205
6. Fischer T, Krauss C. Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research 2018; 270 (2) : 654–669
7. Yoshihara A, Fujikawa K, Seki K, Uehara, K. Predicting stock market trends by recurrent deep neural networks. European Journal of Operational Research 2018; 270 (2) : 654–669
8. Hu Z, Liu W, Bian J, Liu X, Liu T. Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction. Proceedings of the eleventh ACM international conference on web search and data mining 20 2018; 261–269

9. Chen J, Tsai Y. Encoding candlesticks as images for pattern classification using convolutional neural networks. Financial Innovation 2020; 6 : 1–19
10. Zhang H, Goodfellow I, Metaxas D, Odena A. Self-attention generative adversarial networks. International Confer4 ence on Machine Learning 2019; 7354–7363
11. Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434 2015.
12. Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y. Generative 8 adversarial nets. Advances in neural information processing systems 2014; 2672–2680