Based On The Hybrid Model Of XGBoost Applied Research In Stock Forecasting

The stock market is one of the important components of a country's economic market.Stocks are actually the most important way for companies to raise working capital.With the development of the socialist market economy with Chinese characteristics,not only investors are always concerned about stocks,but ordinary people also regard it as a channel for investment and financial management.With the continuous advancement of the times,people's living standards have been increasing.In addition to solving the problem of food and clothing,there is surplus money available for investment.More and more people are turning their attention to stock market investment,which provides financial conditions for the development of the stock market.However,in the complicated stock market,how to find the optimal stock has become an urgent problem to be solved.This is not only a unilateral confusion for investors,but also a focus of scholars in the field of stock price forecasting.Therefore,the design and implementation of the stock market forecasting system not only has profound theoretical significance,but also has very important use value.In recent years,due to the rapid development of artificial intelligence,the development of machine learning theory has been promoted,and it has been widely used in various practical applications,which has ignited a wave of machine learning.Based on machine learning theory,this paper explores the application of XGBoost and ARIMA hybrid models in stock forecasting.Then based on the XGBoost and ARIMA models,combined with the theory of machine learning,the corresponding model structure improvement and optimization are proposed,and the corresponding model comparison is made.The main research and innovations of this paper include the following:(1)The XGBoost financial prediction model(GS-XGBoost)optimized by grid search algorithm is proposed.First,according to the idea of ??grid search algorithm,first set the parameter combination interval to be selected.Based on the Xgboost algorithm,in the process of parameter optimization,combined with the idea of ??grid search algorithm,the model is continuously trained,and each function is evaluated by the evaluation function.The classification results of each parameter combination are evaluated,and finally the optimal parameter combination is obtained.Finally,the optimal parameter combination is substituted into the Xgboost algorithm,thereby improving the prediction performance.(2)Propose a stock price prediction model(DWT-GS-XGBoost)combining discrete wavelet transform and optimized XGBoost algorithm.Considering the advantages of DWTand XGBoost models comprehensively,discrete wavelet transform is used for data denoising and decomposition,and then the XGBoost model optimized by grid search is used to train and predict the processed stock data set,and predict with GS-XGBoost model.The results were compared and analyzed.The experimental prediction results show that the prediction effect of DWT-GS-XGBoost model is better than that of GS-XGBoost model.(3)A hybrid model of discrete wavelet transform,ARIMA and optimized XGBoost(DWT-ARIMA-GSXGB)is proposed to solve the stock price prediction problem.The experimental results show that the DWT-ARIMA-GSXGB stock price prediction model has good fitting ability and generalization ability,which greatly improves the prediction performance of a single ARIMA model or a single XGBoost model in predicting stock prices.