Modelling the dependence structure between bitcoin and major currencies using copula dynamic during extreme period
This article studies the bivariate dependence structure for six pairs of daily returns Bitcoin with Euro (EURO), Pound sterling (GBP), Japan yen (JPY), Canadian dollar (CAD), Australian dollar (AUD) and Chinese renminbi (CNY) from 1 January 2017 until 31 December 2019. A time-varying copula approach...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Penerbit Universiti Kebangsaan Malaysia
2025
|
| Online Access: | http://journalarticle.ukm.my/25774/ http://journalarticle.ukm.my/25774/1/149-165%20-.pdf |
| Summary: | This article studies the bivariate dependence structure for six pairs of daily returns Bitcoin with Euro (EURO), Pound sterling (GBP), Japan yen (JPY), Canadian dollar (CAD), Australian dollar (AUD) and Chinese renminbi (CNY) from 1 January 2017 until 31 December 2019. A time-varying copula approach is employed to explore the dependence between Bitcoin and major currencies during extreme period. The Autoregressive-Generalized Autoregressive Conditional Heteroscedastic-t (AR-GARCH-t) model is applied to estimate the marginal distributions whereas Gaussian and Symmetric Joe-Clayton (SJC) copula models are used to analyse the joint distributions. The results showed there is no tail dependence for all pairs. Overall, Bitcoin has a strong dependence with GBP and lowest dependence with CNY within the pairs using time-varying Gaussian. However, the overall time-varying dependency are very low, which indicates that Bitcoin can acts as a hedge asset against the risk of currency market. Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) have been employed in selection of the best model and conclude that time-varying Gaussian copula is the most appropriate method in measuring the dependence between the currencies. |
|---|