The Research Of The Pricing Model Of Oil Futures

Crude oil futures refer to the futures of the items whose forward crude oil price is the target.It is the most significant oil futures.The strategic characteristics of oil even decide the status of the country behind its power to determine rates.However,the power of discourse on oil prices is monopolized by countries with developed oil futures markets such as the United States and the United Kingdom.With the development and improvement of China’s financial market,China established the Shanghai Fuel Oil Futures Market in 2008,strengthening the power of discourse has become a vital requirement for improving China’s international statusThis paper draws on the research results of crude oil futures pricing and machine learning at home and abroad,studies related literature and uses the WIND terminal crude oil futures data from March 27,2018,to December 31,2019.The machine learning is used to predict the yield of crude oil futures for pricing,and the prediction effects of three linear models and SVR models are analyzed.Then we use the Stacking model integration technology integrate the basic model.Finally,the relevant countermeasures for the development of the crude oil futures market are put forward.In this paper,the relatively new crude oil futures were selected as research subjects.During the process,a large number of factors were used in many models for prediction.It was found that the models of Ridge,Lasso,ElasticNet and SVR after the machine learning had different advantages in the yield prediction.All of them have a certain significance for improving the pricing model of oil futures.At the same time,from the Stacking integration,it was found that new model had a better performance of steadily improving the accuracy rate than the basic model,which is the innovation of this paper.Oil futures have been developed for a short time in China and there are few related researches of machine learning,so the research content of this paper will be of certain significances.