Assessing Irish Banking Stocks through Time-Series Forecasting and Quantitative Trading Models

Shaik, Sadhik (2024) Assessing Irish Banking Stocks through Time-Series Forecasting and Quantitative Trading Models. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (419kB) | Preview

Abstract

This study investigates the application of predictive modelling techniques for stock price forecasting of AIB and BOI banks, Comparing the performance of ARIMA, Linear Regression, LSTM, and a Hybrid ARIMA-LSTM model. The methodology includes comprehensive data preprocessing, feature engineering, and evaluation using metrics such as RMSE and R². Results demonstrate the potential of the Hybrid ARIMA-LSTM model to leverage the strengths of statistical and deep learning approaches, effectively capturing both linear trends and non-linear patterns in stock price data. While Linear Regression and ARIMA excel in simplicity and accuracy for specific datasets, the Hybrid ARIMA-LSTM showcases superior adaptability to complex and volatile data structures with an accuracy of 99%. The predictive capability of model is narrow with unseen smaller samples. The study finds that BOI stock has stable and reliable daily returns compared to AIB. This work underlines the importance of the selection of predictive models with respect to data characteristics for financial forecasting and decision-making in stock markets. However, the research also recognizes several limitations, including the impact of external economic factors and market sentiments on stock prices. The work may be extended by incorporating those external variables, exploring newer ensemble learning techniques, or assessing real-world applicability in dynamic market scenarios. This research provides valuable findings upon differentiating model results on the amount of data we use for model training is significant, that could be useful for both academics and financial practitioners in the development of robust methodologies for predicting stock prices.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Milosavljevic, Vladimir UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science H Social Sciences > HG Finance > Banking H Social Sciences > HG Finance > Investment Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	04 Sep 2025 14:32
Last Modified:	04 Sep 2025 14:32
URI:	https://norma.ncirl.ie/id/eprint/8799

Actions (login required)

View Item