Industry asset allocation model based on LSTM neural network

The rapid development of China's financial market in recent years has brought not only convenience but also challenges to investors. How to effectively allocate assets is one of the problems that investors need to solve. The Black-Litterman model not only solves the traditional mean variance model parameter-sensitive issues, but also allows investors to add investment perspectives to the model. It is a closely watched asset allocation model. However, investors may not be able to give a suitable investment perspective because of their inexperience, they cannot play the application value of the model. This article uses a long-term short-term memory (LSTM) neural network to express quantitative views to solve this problem. As a numerical example, we use ShenWan first-level industry index as an asset pool to build a portfolio, the result of the example shows that the asset allocation model constructed in this paper has a higher Sharpe ratio and annualized rate of return than other reference models. © 2021, Editorial Board of Journal of Systems Engineering Society of China. All right reserved.