Stock market crashes modeling: stochastic cusp catastrophe application

We show that the cusp catastrophe model explains the crash of stock exchanges much better than other models. On the data of U.S. stock markets we demonstrate that the crash of 1987 may be better explained by cusp catastrophe theory, which is not true for the crash of 2001. With the help of sentiment measures, such as index put/call options ratio and volume (the former models the proportion of the chartists, while the latter the fundamentalists), we have found that the 1987 returns are clearly bimodal and contain bifurcation flags. The cusp catastrophe model fits these data better then alternative models. Therefore we may say that the crash may have been led by internal forces. However, the causes for the crash of Sept. 11, 2001 are external, which is also evident in much weaker presence of bifurcations in the data. Thus alterantive models may be used for its explanation.