Stock prediction and trading using low-shot machine learning

• Main Author: Tee, Chee Zhang
• Other Authors: Wang Lipo
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/138830
• Institution: Nanyang Technological University

Many economists, researchers and analysts find it difficult to predicting stock prices. In fact, investors are keen to know the potential of the stock price prediction in the research field. Lots of investors are eager to determine the future stock market trend for a safe and profitable investment. In this paper, long short-term memory (LSTM) network is used with the implementation of low-shot learning to predict the future price of stock. By using the method mentioned to determine the Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) between the predicted and actual stock price value, the experimental results indicate that the proposed method outperforms other benchmark model, and that the optimal sub-sampling rate and various parameters can predict the recent stock price of Standard and Poor 500 (S&P 500) accurately.