Research on Forecast Model Based on BP Neural Network Algorithm

Abstract

Back propagation Neural Network is a well-known method of prediction. Due to the development of the neural network technology, there has been an increasing amount of neural network application research in the last decade, with a majority of application focusing on financial decisions. Firstly, this paper demonstrates the algorithm of back propagation Neural Network. Then, the application of Back propagation algorithm in prediction sector is explained in detail through the design of pre-warning model of financial distress for real estate industry in China. Meanwhile Z-score model is utilized to produce the only output nodes of Back propagation Neural Network. The research can provide evidence of forecast performance based on BP Neural network technique.