Not logged in.
Quick Search - Contribution
Contribution Details
| Type | Bachelor's Thesis |
| Scope | Discipline-based scholarship |
| Title | Testing Bitcoin's Bubble Behavior in the Period 2016-2018 |
| Organization Unit | |
| Authors |
|
| Supervisors |
|
| Language |
|
| Institution | University of Zurich |
| Faculty | Faculty of Business, Economics and Informatics |
| Number of Pages | 44 |
| Date | 2020 |
| Zusammenfassung | In this paper price returns of Bitcoin are evaluated within the periods of 2016 until 2018. The thesis will specify on super exponential growth patterns and test Bitcoin’s behavior on financial bubbles. In order to proceed the Log Periodic Power Law Singularity (LPPLS) Confidence TM multi scale indicators model of Professor Sornette is used. He is globally well-known Professor of the “Eidgenösisches Technische Hochschule” (ETH) and being an expert in the field of financial bubbles, he predicted several of them, as for example the subprime Mortgage bubble. The purpose of this paper will be to first identify potential bubbles and then to seek for indicators which would help to predict the existence of a bubble. Professor Sornette and some colleagues already made an approach on testing Bitcoin for bubble behavior, though for a slightly different period. Secondly the question whether an active trading strategy would outperform a comparable passive trading strategy will be answered by using a similar trading strategy, as presented by Greenwood, Shleifer, and You (2019). In chapter 2 an extended methodology of Johansen and Sornette (1998), will be applied in order to identify bubble peaks. Those will be categorized into long bubbles and short bubbles, accordingly to the definition of Gerlach, Demos, and Sornette (2019). Furthermore, by using the Lagrange Regularisation Approach implemented by Demos and Sornette (2017), the period before a peak can be limited to its minimum and defined as the beginning of a bubble phase. Having an overlapping period, regarding the paper of Gerlach, Demos, and Sornette (2019), will help applying the Epsilon Drawdown Method correctly. In order to diagnose the period of 2016-2018 for bubble behavior, the modification of the LPPLS ConfidenceTM multi scale indicators model, implemented by Filimonov, Demos, and Sornette (2017), is used. The LPPLS calibration should provide an ex-ante prediction, for the in section 2.1 identified bubble peaks. In chapter 3, a summary of the Rational Expectation Bubble model is given. A basic framework, explaining the nature of a bubble and why bubbles keep happening. Moreover, a modification of the LPPLS ConfidenceTM multi scale indicators model will be presented. Given the start and the peak of a bubble regime is known, it is possible to predict the approximate day of the burst of the bubble. The modification works only post-mortem, thus not being used to answer the hypothesis, of whether it would have been possible to predict the burst of the bubble. In chapter 4, a little twist will be given, by elaborating a trading strategy on the base of Greenwood, Shleifer, and You (2019). The idea is to test, whether an active trading strategy would outperform a buy and hold strategy. The trading strategy will be tested, within a bubble phase itself and for the whole Period of 2016-2018. The indicators observed in the LPPLS calibration, will be implemented into the trading strategy as selling signals. Meaning, when strong signs of bubble behavior are found, the holdings of an investor are shifted into a market- or risk-free portfolio. In chapter 5, the findings will be discussed, regarding their correctness and relevance. The explanations will be based on own thoughts and some relevant papers. Furthermore, the intention is to give some ideas for future improvements and further investigations. |
| Export |
BibTeX
|