This repo contains the code for research paper on the speculative bubbles in ethereum’s price. Jonathon Mckeown and I started this research as a group project in our Senior Sophister undergraduate year at Trinity College Dublin and later published it in the Student Economic Review. To view the full paper, please see: https://www.tcd.ie/Economics/assets/pdf/SER/2020/17emprical_investigation_into_speculative_bubbles.pdf
Cryptocurrencies have dramatically changed the way we think about money. With the rapid development of electronic payments around the globe, their popularity has skyrocketed over the past decade and cryptocurrencies now occupy a considerable space in many investors' portfolios. With the raise in popularity has come a huge amount of volatility, and consequently most of these cryptocurrencies are prone to mildly explosive speculative bubbles.
This paper seeks to establish whether the Ethereum market reflects periodically collapsing speculation in its prices, using a recursive unit root procedure (Backward Supremum Augmented Dickey Fuller) introduced by Phillips (2013a).
Our results strongly support the hypothesis of multiple bubbles emerging in the series. We date-stamp 15 bubbles between July 2010 and February 2014, ranging from 1 day to 73 days in length. The longest bubble period, spanning between February 2, 2017 and April 17, 2017, is discussed with respect to price evolution in the Bitcoin market, as well as with regards to Ethereum software being adopted for commercial use.