Identifying price bubble periods in the Bitcoin market-based on GSADF model

Abstract

The advert of cryptocurrency era is signified by the luanch of Bitcoin, which has been considered as a milestone event. Ever since its emergence, Bitcoin has experienced rapid price increase along with great price fluctuations. This paper aims to explore the beginning and the culmination of possible speculative bubbles the Bitcoin market. Based on the daily closing price of Bitstamp, a major Bitcoin exchange platform, from September 13, 2011 to October 12, 2020, this paper adopts GSADF (generalized sup augmented Dickey–Fuller) to explore the existence of the bubble in the Bitcoin market, estimate the time of emergence and collapse of the bubble, investigate the causes of the bubbles based on its time order, and attempts to put forward some policy suggestions. Through this study, we have found that: (1) There are a lot of speculative bubbles. During the research period, there were a total of four long bubbles (i.e. 2013.01.25–2013.04.05; 2013.11.05–2013.12.10; 2017.04.27–2017.07.05; 2017.07.15\u2009–2018.01.09) and four short bubbles (2016.06.08–2016.06.18; 2016.12.30–2017.01.15; 2017.02.22\u2009–2017.03.07; 2019.06.19\u2009–2019.06.25). (2) The emergence and burst of bubbles were mostly related to the occurrence of relevant events and policies. (3) Bitcoin market is far from mature. A key policy risk triggering the speculative behavior is policy announcement. The prominent innovation of this paper is that the latest developed bubble detection method GSADF has been employed in the Bitcoin market. This method can detect the existence of multiple Bitcoin bubbles and their durations.