How well the log periodic power law works in an emerging stock market?

ABSTRACT

A growing body of research work on Log Periodic Power Law (LPPL) tries to predict market bubbles and crashes. Mostly, the fitment parameters remain confined within certain specific ranges. This paper examines these claims and the robustness of the reformulated LPPL model of Filimonov & Sornette (2013) for capturing large falls in the S&P BSE Sensex, an Indian heavyweight index over the period 2000–2019. Thirty-five mid to large-sized crashes are identified during this period, forming a clear LPPL signature. This confirms the possibility to predict the embedded risk of future uncertain events in the Indian stock market with the LPPL approach.

KEYWORDS:

• Financial bubbles
• crashes
• log-periodic power law
• fit method

JEL CLASSIFICATION:

• C13
• C53
• G01

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The BSE lists close to 6,000 companies and is one of the largest exchanges in the world, while Sensex measures the performance of the 30 largest, most liquid and financially sound companies across key sectors of the Indian economy.

² The data that support the findings of this study are available from the corresponding author upon reasonable request.

³ Drawdown is the cumulative loss from one local maximum to the immediate next minimum; a size that is above the threshold ‘ε’.