Enhancing Investment Recommendations with Temporal Learning Models: A Study on Stock and ETF Portfolios

Teixeira de Lima, Caio Cesar (2025) Enhancing Investment Recommendations with Temporal Learning Models: A Study on Stock and ETF Portfolios. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (2MB) | Preview
Preview	PDF (Configuration Manual) Download (759kB) | Preview

Abstract

The use of technologies and application of artificial intelligence in investment management has advanced significantly in recent years, but we still used legacy systems, often assume static investor profiles, ignoring temporal changes in behavior and market conditions. This research proposes the use of temporal models such as LSTMs and HMMs, to develop a system of dynamic investment recommendations based on historical data on stocks and ETFs, seeking to fill a gap in the continuous adaptation of portfolios and the validation of their financial impact. The motivation arises from the need for tools that respond to stock market volatility and offer operational accuracy to investors in an automated and controllable way. The methodology includes time series analysis, model training, and validation with financial metrics in simulated portfolios. It is expected to contribute to an accessible and scalable framework, providing adaptive recommendations and performance evidence.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Jameel Syed, Muslim UNSPECIFIED
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HG Finance > Investment
Divisions:	School of Computing > Master of Science in Artificial Intelligence for Business
Depositing User:	Ciara O'Brien
Date Deposited:	24 Jun 2026 11:53
Last Modified:	24 Jun 2026 11:53
URI:	https://norma.ncirl.ie/id/eprint/9406

Actions (login required)

View Item