Are bitcoin investors overconfident? A FIEGARCH approach

Abstract

We propose in this article to study the behavior of investors in the bitcoin market in order to test whether investors' overconfidence is a driver of excess volatility, often associated with the aforementioned market. This paper presents an attempt to deepen the previously published studies by adopting a new ARMA(p,q)-FIEGARCH(1,d,k,1) parametrization capable of capturing the overconfidence element as well as simultaneously accounting for possible long memory effect. The data used in this study consists of daily closing prices along with daily exchange volume of Bitcoin, spanning the period ranging from 01/01/2012 up to 31/05/2018. The results and conclusions drafted in this research paper could help to understand the formation of volatility in the Bitcoin market. Therefore, this kind of studies will enable investors to better predict bubbles and irrational exuberances. The main contribution of the present article is drawn from the broadening of previous studies by adopting a newly constructed model, which combines capturing asymmetric response, long memory along with the overconfidence element.