Hybrid Prediction Based on BP Neural Network and Markov Chain

摘要

At present, there are many kinds of prediction methods, how to play the advantages of different prediction methods and take advantage of useful information contained in the data have become the focus of research. This paper starts from the studies of combination prediction method. First, it described the current research status of combination prediction methods. Second, it analyzed the characteristics and applicability of different prediction methods. Finally, it proposed a hybrid prediction method based on BP neural network and Markov chain, and applied it to the prediction of personal payment in endowment insurance fund. The experimental results show that the proposed method is effective and feasible.