Stock price monitoring system

Ng, Chun Ming (2024) Stock price monitoring system. Final Year Project, UTAR.

Preview

PDF Download (3260Kb) | Preview

Abstract

This project is a development-based project with deep learning approach for academic purpose. This stock price monitoring system is developed to help users keep track of stock price movements and predict the closing price, so that the users can reap benefits from it and make profits. Algorithms trading is commonly practiced in the financial market due to high accuracy and least human error. Nevertheless, there are few issues occur in the Malaysia stock market, which include absence of stock price prediction function for Bursa listed stocks, poor data visualization and lack of investment recommendation. The main objectives of this project are to to develop a stock price forecasting model, to build a dashboard to present data, and to provide investment recommendations. As stock price is time series data, a time series prediction algorithm is being utilized to build a deep learning model, namely Long Short-Term Memory (LSTM). Consequently, Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE) are used to evaluate the performance of the prediction algorithms. The methodology used in this project is Cross-Industry Standard Process for Data Mining (CRISP-DM), which is a common standard for data mining projects. Lastly, the expected outcome of this project is to have a web application to be built on Streamlit, that incorporates the deep learning model of stock price prediction and displays a dashboard with stock price insights. The stocks of this project are referring to the stocks in Bursa Malaysia, the one and only stock exchange of Malaysia. With the presence of this proposed system, it can benefit the retail investors to make investment decisions by predicting future stock price.

Actions (login required)