Author: Siyabulela Monde Mathe (ST10114635) Date: January 2024 Institution: Varsity College - Newlands
Description:
This research investigates the comparative performance of various machine learning algorithms in the development of trading bots. It addresses the gap in research regarding the effectiveness of different algorithms in financial trading environments. The study explores traditional financial analysis approaches, their limitations, and the advantages of machine learning in automated trading. It aims to identify the most effective algorithms for trading bot development, considering factors such as data quality, algorithm selection, and trading strategies.
Key Components:
- Declaration: Statement of original work.
- Abstract: Overview of the research, its objectives, and potential impact.
- Acknowledgements: Appreciation for contributors.
- Table of Contents: Outline of the thesis structure.
- Chapter 1: Introduction:
- Background and aims of price prediction.
- Thesis structure overview.
- Contextualization of machine learning in trading.
- Research questions, aims, and objectives.
- Problem statement, assumptions, and rationale.
- Chapter summary.
- Chapter 2: Literature Review:
- Introduction to machine learning in trading.
- Time series analysis and traditional financial analysis approaches.
- Drawbacks of existing quantitative finance works.
- Structured vs. unstructured data.
- Supervised and unsupervised learning.
- AI, ML, and DL concepts and their applications in trading.
- Algorithmic trading, market conditions, and technical/fundamental analysis.
- Sentiment analysis.
- Chapter summary.
- Chapter 3: Methodology:
- Research method, approach, and design.
- Data collection and analysis methods.
- Ethical considerations and study limitations.
- Algorithms used.
- Chapter summary.
- Chapter 4: Data Analysis:
- Exploratory data analysis.
- Data pre-processing and preparation.
- Model implementation and evaluation.
- Training and test set evaluation.
- Chapter 5: Future Research and Conclusion:
- Findings and discussion.
- Conclusion and recommendations for future research.
- Chapter 6: References:
- List of cited works.
- Figures and Tables: Visual representations of data and results.
Purpose:
This research contributes to the understanding of machine learning's role in quantitative financial analysis and trading bot performance. It provides insights into selecting and developing effective trading algorithms, benefiting researchers, FinTech stakeholders, investors, and academic disciplines.