Stock Index Futures Price Forecast Based On Wavelet Kernel Support Vector Machine Regression

This article is based on the wavelet kernel support vector machine regression of stock index futures price forecasting program,the reason why choose to predict the price of stock index futures is mainly affected by the special conditions of the financial market.In 2015,stock index futures suffered the most severe regulation in history.After 2017,regulation and control have been liberalized.However,the stock index futures market still faces the situation of insufficient liquidity under trading volume.However,after the limit of stock index futures,this strategy has been challenged enormously.The trading volume of stock index futures cannot meet the need for full hedging of risks.Therefore,some private equity institutions have shifted to flexible selective timing hedging strategies,while some small and medium financial options have not hedged.Or establish a unilateral position of stock index futures to invest.Under such circumstances,it is very important to accurately predict the price of stock index futures.This paper draws on the construction method of wavelet kernel functions.By giving three kinds of wavelet kernel function pre-calculation matrices,the wavelet basis function is brought into the regression model of support vector machine.The regression model based on wavelet kernel support vector machine is established and then adopted.The genetic algorithm and particle swarm optimization algorithm optimize the parameters of the wavelet kernel support vector machine regression model,and obtain the optimal parameters of the wavelet kernel support vector machine regression,and thus make predictions on the closing prices of the main contract of the three stock index futures on the Chinese market.Finally,the rationality test shows that the prediction performance of the wavelet kernel support vector machine regression is better than the ordinary kernel function.