STOCK MARKET NETWORK ANALYSIS USING MINIMUM SPANNING TREE

• Main Author: MIHDA HAYATI (NIM: 10113067), MILLATI
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/23076
• Institution: Institut Teknologi Bandung

Network can describe interaction between stock markets. Stock market network can be constructed by minimum spanning tree and based on correlation among stock markets. Using return data of 39 stock indices, stock market network is determined in three time periods: before, during, and after global financial crisis 2008. Dynamic changes in network can be seen by centrality scores of each stock indices. There are three centrality measures that are used in this final project. They are degree, betweenness, and closeness centrality. Changes in stock market network explain that there is significant variation of correlation among stock markets through time. Therefore, modeling correlation is important thing to help investors understanding behaviour of stock markets correlation in network. In this final project, <br /> <br /> correlation between stock indices is predicted using DCC GARCH model. DCC GARCH model is applied to data with assumption Gaussian distribution errors.