Auto-Trading Bot

Seow, Yu Xuan (2025) Auto-Trading Bot. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RDS_Seow Yu Xuan_Fulltext.pdf Restricted to Registered users only Download (3MB)

Abstract

This project focuses on developing a comprehensive stock market prediction and trading strategy application, integrating machine learning models with a portfolio management system to assist users in making informed trading decisions across all 11 stock market sectors. The prediction module leverages Linear Regression, Long-short Term Memory (LSTM), and Prophet models to forecast stock prices, with each model evaluated using performance metrics of Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and R-squared to ensure accuracy. The trading strategy module applies key technical indicators like the Relative Strength Index (RSI), Exponential Moving Average (EMA), and Supertrend to generate actionable buy/sell signals, while a backtesting feature enables assessment of the strategy’s effectiveness on historical data. The portfolio module stores and tracks all buy/sell transactions, offering users a detailed performance summary for each stock traded, and helping them monitor profit and loss across various trades. Built using Streamlit, the application provides an intuitive interface, allowing users to input and validate stock tickers, compare multiple stocks, and visualise performance through dynamic charts. While primarily focused on individual stock performance, the system's portfolio module offers key insights for managing trades, with plans for expanding its functionality to support broader portfolio analysis in future iterations.