A STUDY OF NEURAL NETWORK APPLICATION IN BEATING STOCK INDICES

Abstract

Neural networks have been applied to financial applications more and more recently, such as pricing securities, risk management and constructing portfolios. Compared with standard financial methods, neural network can take all the data relevant into account and it can learn not only linear relationships but more complex features of the input data. Another advantage of neural network is that it is more easy to reduce over-fitting and improve the performance on the test set. In this study, we focus on the application of neural networks in constructing portfolios of stocks which can outperform the stock index by a specified level. The results show that the deep portfolios do outperform the Hang Seng index and Shanghai Stock Exchange 180 index. The study also shows that we can make profits by longing the portfolio and shorting the exchanged traded fund (ETF) of index.