Font Size: a A A

Design And Implementation Of The Back-testing Platform For Quantitative Stock Selection

Posted on:2018-12-28Degree:MasterType:Thesis
Country:ChinaCandidate:L X KongFull Text:PDF
GTID:2348330515969847Subject:Software engineering
Abstract/Summary:
Recently with the introduction of relevant theories in the field of financial engineering,the domestic financial industry has made great progress in quantitative investment.Meanwhile,the advantages of quantitative investment are more obvious,in the context of financial market becoming more and more mature gradually and increasingly strict supervision of China’s securities regulators.With the growing sophistication of information technology,to check the stock selection model applied to quantitative investment funds,the usage of traditional methods of manual inspection still leads to much difficulty and greater workload.Therefore,it’s a wise choice to solve such problem in the process of the quantitative investment with computer technology,and quantitative stock selection platform springs up.Through the acquisition and analysis of the demand from the relevant employees and some investors of the securities fund,the traditional business processes of the investment fund have involved detailed and planed research.Finally the end user and major functions of the platform are confirmed with the software requirement specification.Specific functions of the system are as follows:1.Data processing,designed to store and constantly update price data and the fundamental information of A-share stocks,which also provides proper access to the above data.2.Backtest process,as far as possible to restore the real terms required by the actual transaction process and backtest the strategy to check its performance with historical data.3.Simulation transaction,to test selection stock strategies in simulation situation close to the real with some trading interfaces provided by Wind? application software.4.The result part shows some relevant data used to show the backtest process,while the attribution analysis uses some corresponding models to analyze strategies to figure out the influencing factors leading to their performance.In the design and implementation,taking other popular quantitative investment platforms for reference and its own characteristics in consider,this paper develops a set of feasible technical solutions.Taking into account the convenience of the rapid implementation of the strategy,the backtest platform of quantitative stock selection early chooses MATLAB as the major program language.While in the data processing for performance requirements,some more low-level parts are accomplished with Java which are called with the given interfaces.In terms of storage and access capabilities,MySQL the open source database management system could satisfy the target system.The backtest platform of quantitative stock selection after functional and performance testing,has meet the requirements of the functional and quality requirements.In the actual use process,the platform can provide investors with the performance of the stock selection strategy model in retrospective historical data,generate the analysis report of attribution analysis according to the selected evaluation model,and show important information for them to make investment decisions.
Keywords/Search Tags:Quantitative stock selection, Backtest, Attribution analysis, MATLAB
Related items