Modelling Stock Market Crashes: The Case of Bucharest Stock Exchange

Abstract In this paper we investigate the bubble behaviour of Bucharest Stock Exchange, using log periodic power laws models. Analysing the behaviour of the most speculative index from Bucharest Stock Exchange, BET-FI, we prove that LPPL models are a useful tool in recognizing the behaviour of a stock market bubble and they have predictive abilities for the critical point of a bubble. Iterative calibration of the model for BET-FI regime led to a reasonable estimate of the stock market crash in January 2008. Using the same methodology, the anti-bubble regime from 2008 is fitted and we find an accurate prediction of the local point of phase transition from 27/10/2008.