266. DCC-MGARCH AND WAVELET COHERENCE ANALYSIS IN EXPLORING THE CONTAGION EFFECT BETWEEN CRYPTOCURRENCIES AND FIAT CURRENCIES
Invited abstract in session TA-9: Statistical methods for finance, stream OR in Finance and Insurance .
Tuesday, 8:30-10:00Room: Clarendon SR 2.01
Authors (first author is the speaker)
| 1. | Tea Šestanović
|
| Department of Quantitative Methods, University of Split | |
| 2. | Josip Arneric
|
| Faculty of Economics and Business, University of Zagreb | |
| 3. | Klara Luketa
|
| OTP bank d.d. |
Abstract
This paper investigates the phenomenon of financial contagion between cryptocurrencies (Bitcoin, Ethereum) and fiat currencies (USD, EUR, GBP). Dynamic Conditional Correlation (DCC) Multivariate Generalized Autoregressive Conditional Heteroscedasticity (MGARCH) model is estimated for each pair of currencies, i.e. crypto vs fiat, since it provides time-varying conditional correlations based on which the properties of currencies as diversifiers, hedge or safe heaven can be examined. Kolmogorov-Smirnov (KS) test is used to investigate the contagion effect by comparing the distribution of conditional correlations in pre-crisis and crisis period. However, source of contagion can not be extracted from these methods. Therefore, wavelet coherence analysis (WCA) is used to elaborate on the lead-lag relationship between the pairs of currencies. The results show that before the pandemic, Bitcoin and Ethereum had a weak negative correlation with USD, acting as a hedge, while during the pandemic, they become safer havens. Correlation with EUR remains close to zero, positioning them as a hedge. In relation to GBP, they exhibited weak positive correlations, making them diversifiers. KS tests revealed the dominance of the crisis over the pre-crisis period for the pairs EUR-ETH, GBP-BTC, and GBP-ETH, indicating the presence of contagion. WCA confirms the correlations between cryptocurrencies and fiat currencies, enabling the lead-lag relationship analysis for each period.
Keywords
- Economic Modeling
- Finance and Banking
- Financial Modelling
Status: accepted
Back to the list of papers