Research On Visual Analysis For The Comparison And Evaluation Of Quantitative Timing Strategies

In the process of stock investment and trading,both institutional investors and individual investors face a common problem: how to choose stocks to buy or sell,to obtain excess returns.This is the timing problem in the stock investment.Around this issue,many scholars and financial market researchers have conducted a lot of research and analysis,and published hundreds of quantitative timing strategies,using past market data for back testing to verify the effectiveness of the strategies.However,the lack of effective tools to manage these strategies hinders the average investor from picking an accurate quantitative timing strategy.On this basis,this paper introduces visual analytics technology and investigates new visual analytics methods,focusing on the problem of comparing and evaluating quantitative timing strategies,and proposes a novel visual analytics method for quantitative timing strategy comparison and evaluation to provide effective support for investors’ decision making.In view of the above analysis,this paper conducts research on the visual analysis method for quantitative timing strategy comparison and evaluation.The main work includes:1.This paper uses quantitative trading models to refine data and proposes a visual analysis framework oriented to the comparison and evaluation of quantitative timing strategies.To reveal the performance of different timing strategies under the same stock trends and within the same time period,as well as the performance of the same timing strategy under different stock trend and within different time periods.It can not only solve the problem of investors comparing multiple quantitative timing strategies at the same time,but also evaluate quantitative timing strategies in multiple dimensions,thus improving the correctness of strategy selection and helping investors to obtain higher returns2.This paper presents a novel visual analysis method for comparing and evaluating quantitative timing strategies by combining circular,sector and curved glyphs and using color coding as well as a dimensionality reduction algorithm to show the performance of each quantitative timing strategy over different time periods.In addition,Comparison View,Trading Signal View,Stock Return View and Candlestick View share the same timeline and add interactive means such as highlighting and linking to assist users in comparing and evaluating the overall performance of quantitative timing strategies in multiple dimensions,providing users with different visualization needs.3.This paper designs and implements a visual analysis system to help users explore quantitative timing strategies comprehensively.The overall framework of the system adopts a modular idea and mainly contains three parts: data layer,business layer and application layer,which improves the ease of use and maintainability of the system.This paper verifies the usefulness and effectiveness of the system in this paper from the practical application level.Through two case studies with investment experts,it is verified that the system in this paper is advantageous for visual comparison of multiple timing strategies and comprehensive evaluation of timing strategies.In addition,this study invites in-depth user studies with investigators from different research backgrounds to validate the usefulness and usability of our approach through qualitative and quantitative evaluations.