Stock Price Forecasting Based On Support Vector Machine And GARCH Models

Time series analysis based on stochastic process theory and statistical methods studies the statistical laws of data sets to solve the problems in real life.Now it has been widely used in various industries.Stock price is an obvious time series in financial market.Support vector machine regression model and generalized autoregressive conditional heteroscedasticity model are common models to deal with time series,which have good effect on the simulation and prediction of financial market data.With the progress of society,people's living standards have improved year by year,people pay more and more attention to the stock financial market,pay attention to the stock price fluctuations.Therefore,the prediction of stock price has important practical significance for the majority of shareholders,investors and enterprise financing.This paper is divided into five parts.Firstly,the first chapter introduces the existing financial stock market forecasting methods on the basis of a large number of references.In the second chapter,the theoretical basis of support vector machine regression model is introduced in detail,and the kernel function used in the model is explained.The third chapter introduces the theoretical basis of the generalized autoregressive conditional heteroscedasticity model.In the fourth chapter,taking the daily and minute stock price data of SF holdings in Shanghai and Shenzhen A-shares from 2017 to 2019 as the training data set,we determine the model parameters of these obtained stock price data series,and then fit the support vector machine regression model and GARCH model,compare the prediction accuracy of the two models,and judge that the two models can better predict the stock price,This paper analyzes the advantages and disadvantages of the two models used in stock price series fitting,and makes a reasonable prediction analysis of the future stock price.The fifth chapter is the result of this paper.Through the analysis of the data in the fourth chapter,it points out that both support vector machine regression and GARCH model can well predict the time series model,and compares the advantages and disadvantages of the two models.This paper mainly through the use of support vector machine regression and GARCH model to predict and analyze the stock,hoping to provide some theoretical basis and reference value for investors in the financial industry and enterprise financing.