Study On Financial Change Point Of Impact Of Public Health Events On Various Industries

The sudden occurrence of public health emergencies is unknown and uncertain,which not only damages people’s lives and health,but also impacts the economy of countries and regions.The COVID-19 epidemic is a typical case of public health emergencies.At the end of 2019,the COVID-19 epidemic broke out in Wuhan and quickly spread to all parts of the country.This epidemic not only severely damaged the lives and health of the Chinese,but also caused a certain degree of impact on the Chinese economy by the measures taken to prevent and control the epidemic,such as suspension of work and production,traffic control and so on.Studying the economic conditions and change points of different industries under the epidemic can provide economic enlightenment for public health emergencies and make anticipatory preparations for possible changes in the economy.This paper studies the economic conditions and change points of some industries under the epidemic,One leading stock in each of the six industries is selected as the research object.The main research contents are given as follows:The first part studies the economic conditions of the six industries.The SVR problem is transformed into an optimization problem to solve.The Gaussian kernel function is used as the kernel function in the SVR model,and the simulated annealing particle swarm optimization algorithm is used to optimize the parameters in the SVR problem.The ARIMA,SVR,and ARIMA-SVR models were established for the six stock sequences.Taking root mean square error and average error percentage as the evaluation indexes,the experiment result shows that ARIMA-SVR model achieved better fitting effect than ARIMA model and SVR model.The second part studies the change points of six industries.The multiple change point problem is transformed into a single change point problem by using two piecewise method.The Bayesian estimation formula of change point in ARMA model is derived by using the properties of inverse gamma distribution and t-distribution.The Bayesian method of change point in ARMA model is used to study the change points of six stock sequences,the Bayesian estimate and posterior distribution of the change points are obtained respectively,the time and possible reasons of the change points are derived.