Analysis of stock market based on visibility graph and structure entropy

Abstract

Complex network has been widely used to reveal the rule of complex system. Recently, visibility graph method is useful tool to convert time series into complex network. And that, structure entropy of network well describe property of complex network. In this paper, a new method, based on visibility graph and entropy methods, is proposed for studying stock market. The daily closing price of stock is regarded as time series. And then, stock network are established from the time series by the visibility graph method. Finally, some properties of stock market are described by measuring the change of structure entropy of the stock networks. The proposed method is applied to assess performance of six market indices of six countries (China, America, Germany, France, Japan, Britain) for 28 years. Meanwhile, some virtual stock markets are discussed. The results show that the economic activity such as financial crisis, that affects the change of network structure and the proposed method can effectively describe some properties of stock market.