Research And Application Of Digital Twin System For Trading Market Based On Multi-Agent

It has been more than 400 years since the development of the trading market.In this long historical process,the stock market has experienced many long-term severe fluctuations due to the impact of events.For example,in the early stage of the outbreak of COVID-19,the maximum drawdown of the Shanghai Composite Index reached 15.36%,which is reflected in the long-term growth trend of continuous fluctuations.As the development status of enterprises and the mood of market trading subjects can directly have a great impact on the stock market,it is very important for relevant departments to timely take appropriate policies and measures to carry out macro-control under the influence of emergencies.In order to pre-evaluate the impact of policies on the market,a digital twin system of the stock market is built based on the multi-agent model.Firstly,the market players are classified and described,and then the various events affecting the change of asset prices are simulated by using the compound poisson process.Finally,an asset pricing method based on the assumption of normal distribution is proposed according to the proportion of buyers and sellers.The paper is briefly described as follows:(1)In the pricing design of the model,this paper innovatively introduces the assumption of normal distribution to simulate the proportion of the number of buyers and sellers in the market,and transforms the pricing problem into a quantile calculation problem,and then solves the integral algebraic equation to obtain the asset price fluctuation sequence.(2)In the event shock simulation,the compound poisson process is introduced to describe the event impact,and the event shock intensity is described with a uniform distribution.(3)After the digital twin system is constructed,the different simulation results are statistically tested and analyzed.The results show that the asset prices generated by the digital twin system are subject to geometric brownian motion.What's more,the rate of return and volatility are in line with those of securities market in China.(4)In the applied research on the digital twin system,the ratio of good news to bad news,news intensity,changes in learners' emotions,and the impact of local shocks on the stock market are discussed.These studies can provide decision aids for policy makers and market regulators,as well as tools for asset managers to measure risks,and provide new ideas for related scholars' researches.