Analysis of stock market crises a network theory approach

Περίληψη

In literature, the issue of economic and financial crisis is well studied and many
econometric models have been developed for predicting economic or financial crisis or investigating if a crisis affects significantly the economy and financial markets. However, only a little work has been done on how crisis affects the way that stock markets behave. In this work, we take advantage of the recently introduced Visibility Graph algorithm, a method that maps a time series into a network, and we examine how stock markets responded to specific events. More speciccally, using five minutes data from January 1996 to March 2016 for the S&P100, S&P500 and S&P1000 indexes, we make a quantitative analysis of the indexes behaviour during this period. We found that indexes react immediately in almost all events and in some cases there are even some early warnings. However, the magnitude of the reaction depends on the size of the index and all indexes do not react in the same way under the same circumstances. Finally, the time series of the returns of an index do not demonstrate the same properties as the price time series.