A Network Visualization Approach and Global Stock Market Integration

Abstract

This paper utilizes the daily prices of the market indices in 57 countries across the world from 2nd January 1997 to 30th August 2012 to examine the potential stock market integration structure by applying a network visualization approach (with both MST network and Graph network), we seek to explain the tendency of integration or co-movement among the markets in order to discover the possible dominant market in a geographic region or globally and the network structure movement during extreme financial events. We also aim to measure the centrality of the corresponding network, and to what extent the network approach reveals the robustness of such centrality indicators. We find a clear trend of markets moving closely and we presented such results in a dynamic and visualized manner. Additionally, we proposed a centrality test based on correlation network and it suggests that the US and Hong Kong markets become most central dominant markets in their economic and geographic region, which is consistent with the literatures. But for Europe we found UK, France and Germany are leading jointly in the region unlike previous literatures suggest the UK is the dominant country in the region.