Font Size: a A A

Basis Prediction And Arbitrage Strategy Design Based On Support Vector Machine Granulation

Posted on:2019-03-02Degree:MasterType:Thesis
Country:ChinaCandidate:P Z HaoFull Text:PDF
GTID:2359330548457699Subject:Finance
Abstract/Summary:PDF Full Text Request
Stock index futures on the financial secondary market are not only important financial instruments for price discovery and for the stabilization of price trends.And it is also an important means of hedging and bilateral arbitrage.The use of stock index futures for arbitrage is not only conducive to the stock index futures to play its due role,but also an essential condition for investors to use stock index futures for hedging and risk aversion.In the current arbitrage of stock index futures,the basis of stock index futures is the core element.If we can make advance forecast and judgment on the trend of basis difference,it will greatly improve the efficiency and flexibility of arbitrage.Traditional prediction of time series focuses on the accurate prediction of single values of sequences,which is difficult to obtain effective results because of the fluctuation characteristics of price series.At the same time,there are more potential market assumptions in the traditional underlying spread arbitrage model.And the flexibility is not high enough to adapt to the ever-changing market transactions.Based on this,from the perspective of a more practical application,we using the support vector machine model that has received much attention in recent years,combined with the information granulation method,the fluctuation range of the basis sequence is predicted and constructed,and the Conduct arbitrage trading on stock index futures.Through the fuzzy granulation processing of the basis data and the optimization of the support vector machine model,the fluctuation range of the basis error is finally effectively predicted.Also,applying the predicted value of the basis fluctuation range to the guidance of the basis arbitrage transaction has also obtained higher returns.Therefore,we hope to use this research plan to provide a new reliable and stable investment strategy for stock index futures investors and arbitrage traders,and actively promote the function of the stock index futures market.
Keywords/Search Tags:SVM regression, Fuzzy information granulation, Futures basis range prediction, Basis arbitrage
PDF Full Text Request
Related items