Development of NTU quant AI platform

• Main Author: Elangovan Karthikeyan
• Other Authors: Li Fang
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science
• Online Access: https://hdl.handle.net/10356/181094
• Institution: Nanyang Technological University

This report showcases the development of NTU Quant AI, a quantitative trading platform that enables traders to trade across multiple asset classes, in different markets, using a single platform, while leveraging AI capabilities such as smart routing. It highlights the main motivations for this project, which can be attributed to the rise of algorithmic trading today, increasing expectations of traders such as the flexibility to utilise their own strategies, and limitations in existing trading platforms in the market. Earlier iterations of this project have developed key components of the system such as the portfolio management system (PMS), Order Management System (OMS) and Execution Management System (EMS), which has paved the way for us to further improve the existing features, identify and develop new ones. In this iteration, our focus would be to overhaul poorly functioning frontend components, alongside the development of an improved strategy pipeline, an improved data collection system through the development of an Extract-Transform-Load (ETL) pipeline and a manual imputation system for users to test their strategies with. The report goes beyond the system analysis of these components, and their technical implementation, to touch upon other essential concepts in software engineering, such as project management approaches we have undertaken. It also elaborates upon the various project resources we have utilised in the development of this platform, which would greatly benefit any reader by enabling them to gain a better understanding of our platform. The report also highlights the author’s focus on the data collection module, elaborating in detail his efforts into the research and analysis work for development of an ETL microservice, and his successful implementation of an end-to-end ETL pipeline for collection of historical news data. The report also touches upon other areas of contributions the author has made, in terms of team-wide contributions, comprehensive testing measures that he has implemented, and future areas of work proposed to expand the platform’s capabilities even further.