Task:
1. Select a supervised algorithm that can predict stock prices basing on historical data.
2. Accordingly formulate a trading strategy (based on predicted values) in order to generate orders dynamically (on same historical training set for back testing) and observe the gain / loss in overall portfolio.
3. Run the same program for 'real-time trades'
4. Select the combination of Machine learning algorithm and Trading strategy to maximize gain for future orders.
Reference:
Andrés Arévalo, Jaime Niño, German Hernández, & Sandoval, J. . (2016). High-Frequency Trading Strategy Based on Deep Neural Networks. International Conference on Intelligent Computing. Springer, Cham.
Yong, B. X. , Rahim, M. R. A. , & Abdullah, A. S. . (2017). A Stock Market Trading System Using Deep Neural Network. Asian Simulation Conference. Springer, Singapore.
Huang, C. Y. (2018). Financial Trading as a Game: A Deep Reinforcement Learning Approach. arXiv preprint arXiv:1807.02787.
Islam, S. R. (2018). A Deep Learning Based Illegal Insider-Trading Detection and Prediction Technique in Stock Market. arXiv preprint arXiv:1807.00939.
Lu, Z. , Long, W. , & Guo, Y. . (2018). Extreme Market Prediction for Trading Signal with Deep Recurrent Neural Network. International Conference on Computational Science. Springer, Cham.
Sezer, O. B. , & Ozbayoglu, A. M. . (2018). Algorithmic financial trading with deep convolutional neural networks: time series to image conversion approach. Applied Soft Computing.
Innovation：
Using deep learning method to deal with algorithmic trading
Combining the prediction of stock price with trading strategies
Truly restoring the trading mechanism of the stock market
Datasets：
Daily stock price from Google Finance and Yahoo! Finance
To-do lists:
Nov 10 - Nov 17: Paper searching, reading, building proper code environment
Nov 18 - Nov 24: Building up codebase and get down to datasets
Nov 25 - Nov 30: Implement ideas
Dec 01 - Dec 07: Check the efficiency and accuracy of the model and fine tune the model，paper writing