Your abstract should be very short - about 50-200 words. It should state:
Pair trading is a simple trading strategy which involves finding a pair of stocks that exhibit similar historical price behavior and then betting on the subsequent convergence of their prices in the event that they diverge. Since this trading style is very applicable to all asset classes, the methods to identify pair trading signals are widely researched. However, the effectiveness of existing rule-based strategies has been reduced because of increased market competition.
In this project, we conducted rigorous studies on existing pair trading strategies proposed in academic papers. We implemented and backtested three popular pairs trading strategies (the Distance method, the Cointegration method with Rolling OLS and the Cointegration method with a Kalman Filter) to better understand their profitability, strengths and weaknesses under different market conditions.
The core achievement of this project was a novel approach to perform pairs trading by adopting a Reinforcement Learning (RL) paradigm. The proposed RL trading agent was backtested over three separate periods (2016, 2017 and 2018) to facilitate a comprehensive comparison between the existing strategies and our RL trading agent.
Finally, we implemented a front-end web application for data visualization of the trade actions and performance of the four pair trading strategies.
[198-word abstract from the final report of the distinguished FYP of 2018-19 group LZ2, Pairs Trading with Machine Learning]