Quantitative trading platform development

This project involves the design, development, and implementation of the NTU Quant AI platform, a web application for quantitative trading. The primary objectives of this project were to establish a strong foundation for future maintainability and extensibility, as well as to develop new features to...

Full description

Saved in:
Bibliographic Details
Main Author: Ng, Tze Sheng
Other Authors: Li Fang
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/174780
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:This project involves the design, development, and implementation of the NTU Quant AI platform, a web application for quantitative trading. The primary objectives of this project were to establish a strong foundation for future maintainability and extensibility, as well as to develop new features to enhance the platform’s offerings. The report outlines the various tools and technologies utilised and highlights the rationale behind their choice and suitability. It continues with an analysis and discussion of the shortcomings of previous implementations and details the solutions employed to address them. Namely, a restructure from a monolithic architecture to a microservices architecture was pursued to facilitate scalability and maintainability concerns. Complementing the architectural changes, a service discovery component was implemented to support dynamic and efficient communication between microservices, enhancing scalability and fault tolerance. Docker was introduced as a means of standardising development environments for all developers to eliminate potential incompatibility and inconsistent code behaviour due to differences between developer machines. Data modelling and storage solutions were also explored to optimise the performance and better fulfil the requirements of the system, adopting the use of a columnar database for financial market data, and a file storage system to overcome previous storage limitations. Aspects of system security were also investigated with options such as password hashing, OAuth 2.0 and Passwordless authentication mechanisms being considered. In closing, the report proposes recommendations for future development efforts and iterations, and reviews the contributions and goals achieved.