Financial study of listed companies using complex network

Abstract

Despite financial market has been developed for hundreds of years, many part of the market are still in black-boxes. We still can not totally understand the market. In this report, we use quantitative approach to examine the interaction of financial markets with different stocks by constructing complex network that includes 86 selected stocks from Hang Seng Index (HSI) of Hong Kong market. In the network being built, the nodes are the stocks and the edges are the connection between stocks. Each edge has the edge weight which is the correlation between a pair of connecting stocks within a window of w-days. We consider the period to be from 29 Dec 2006 to 29 Dec 2009 which is 783 trading days. In this period, the networks are built for all w-day windows at 1 day-to-day time intervals. The window w is a moving window monitoring w days within the whole period. This study considers the effect of the 2007 global financial crisis on the network of Hong Kong market around the time of crisis especially three time periods, before, during and after the crisis. We examine the variation of the network parameters and the quantitative analysis with the change of economic condition. We show that the dynamic of complex network connectivity is highly related to the market. Specifically, different forms of market features are found.