Kaiji Gong

The Extension of the Log Periodic Power Law: How to Predict After the Crash

To judge whether an economic bubble would lead to a financial crash and to estimate the critical time of a crash are significant in financial areas. The Log-Periodic Power Law (LPPL) is an equation that describes how bubbles evolve and grow. By fitting the equation into a financial time series, it is possible to predict the event of a crash. The equation proves to be effective in predicting several financial crises, such as the one in 2008. My research will focus on extending the model to predict market behaviors after the crash. I will adopt some assumptions in LPPL (i.e., imitation behaviors among traders) as well as some more theories from other literatures, and follow the similar track to construct a model that predicts when the market will stop crash. I will then fit the model into financial time series to examine the accuracy of the model.

Message to Sponsor

SURF grants me great honor to become an independent researcher, which is a benchmark in my academic career. I have learned (and I am still learning) how to design research projects, go over various literatures, build up models and examine them by data analysis. I am very grateful that SURF has given me the opportunities to utilize the knowledge Ive learned into real practice. Meanwhile, the fellowship also reveals how graduate school life should be like; one has to be highly self-motivated, keeping asking questions and looking for solutions. SURF encourages me a lot to become a qualified researcher.
  • Major: Mathematics, Statistics & Economics Double Major
  • Sponsor: JSB SURF L&S Fellow
  • Mentor: Robert Edelstein