Deep Neural Network Model Forecasting for Financial and Economic Market

Recently, the Internet financial market has developed rapidly both at home and abroad. Simultaneously, its study has also become the focus of academic circles. The financial markets have higher liquidity and volatility as compared to traditional financial markets. In view of the Internet financial market dynamic (volume and daily trading), it is proposed based on a deep neural network for fusion level time series prediction model. First, the proposed model processes the input of characteristic variables of multiple series (market macrodynamic series and multiseed series) and uses an attention mechanism to fuse the input variables in two dimensions of time and sequence feature. Second, the model also designs an optimization function based on the stability constraints of the prediction sequence, so that the model has better robustness. Finally, a large number of experiments are carried out on real large-scale data sets, and the results fully prove the effectiveness and robustness of the proposed model in the dynamic prediction of the Internet financial market.