Back to Search
Start Over
Statistical analysis of bitcoin during explosive behavior periods
- Source :
- PLoS ONE, PLoS ONE, Vol 14, Iss 3, p e0213919 (2019)
- Publication Year :
- 2019
- Publisher :
- Public Library of Science, 2019.
-
Abstract
- This paper develops the ability of the normal inverse Gaussian distribution (NIG) to fit the returns of bitcoin (BTC). As the first cryptocurrency created, the behavior of this new asset is characterized by great volatility. The lack of a proper definition or classification under existing theory exacerbates this property in such a way that explosive periods followed by a rapid decline have been observed along the series, meaning bubble episodes. By detecting the periods in which a bubble rises and collapses, it is possible to study the statistical properties of such segments. In particular, adjusting a theoretical distribution may help to determine better strategies to hedge against these episodes. The NIG is an appropriate candidate not only because of its heavy-tailed property but also because it has been proven to be closed under convolution, a characteristic that can be implemented to measure multivariate value at risk. Using data on the price of BTC with respect to seven of the main global currencies, the NIG was able to fit every time segment despite the bubble behavior. In the out-of-sample tests, the NIG was proven to have an adjustment similar to that of a generalized hyperbolic (GH) distribution. This result could serve as a starting point for future studies regarding the statistical properties of cryptocurrencies as well as their multivariate distributions.
- Subjects :
- Multivariate statistics
Cryptocurrency
Economics
Science
Materials Science
Test Statistics
Normal Distribution
Social Sciences
Research and Analysis Methods
Normal distribution
Normal-inverse Gaussian distribution
Mathematical and Statistical Techniques
0502 economics and business
Econometrics
Psychology
Humans
050207 economics
Statistical Methods
Materials
Statistical hypothesis testing
Mathematics
Behavior
050208 finance
Multidisciplinary
05 social sciences
Statistics
Commerce
Biology and Life Sciences
Probability Theory
Probability Distribution
Convolution
Models, Economic
Public Finance
Physical Sciences
Money Supply and Banking
Medicine
Probability distribution
Explosives
Volatility (finance)
Volatilization
Mathematical Functions
Value at risk
Finance
Research Article
Statistical Distributions
Subjects
Details
- Language :
- English
- ISSN :
- 19326203
- Volume :
- 14
- Issue :
- 3
- Database :
- OpenAIRE
- Journal :
- PLoS ONE
- Accession number :
- edsair.doi.dedup.....c8a796f9036d50e5ff048d7bff0832c7