Research and Comparison of Random Forests and Neural Networks in Shanghai and Shenzhen Financial 20 Index Prediction

Machine learning has made great achievements in the field o f a rtificial in telligence, es pecially in th e financial industry, which has shown great potential and attracted the attention of the academia and the industry. With the opening of the capital market to the outside world, how to achieve more accurate forecast of the trend of stock index and timely carry out reasonable risk control is one of the important issues concerned by investors. On the basis of literature and theoretical analysis, the public data sets of a well-known securities firm is selected to make predictions of the Shanghai and Shenzhen CSI 20 index based on neural networks and random forest models. The average absolute error (MAE) was then used as a metric to compare the performance of the two machine learning algorithms on the stock index prediction problem. The results show that the model constructed by the random forest algorithm performs better on this problem. Therefore, we believe that making investment decisions based on the results can help investors make effective investment decisions under certain risks.