Financial Secondary Markwet Data Analysis Based On Deep Learning

In the financial secondary market,the data analysis method is mainly based on statistics and artificial modeling methods.Based on the analysis and research of the characteristics of the financial secondary market data,this paper puts forward the use of neural network method in the secondary market Ideas.This paper demonstrates the feasibility of using neural networks in data analysis in the secondary market.And by analyzing the data of financial secondary market,the whole data source is divided into three modules,and according to the characteristics of financial data and priori knowledge of some financial analysis methods,a CNN-LSTM network is designed which is suitable for processing financial data To process the data.The network can analyze the morphological characteristics of the data price and the timing characteristics of the data through the CNN network.Through the combination of the two networks,the data of the financial secondary market can be effectively analyzed.Compared with traditional simple statistical methods and some neural network methods such as Logistic Regression,Convolutional Neural Network(CNN),LongTerm and Short-Term Memory Networks(LSTM),this network compares the forecast of market price changes in a relatively short period of time,The prediction in a long time has a certain improvement,which is about 7% higher than the simple statistical method and about 4% higher than other neural networks.