We use a GARCH dummy model to study the influence of calendar effects on daily conditional returns and volatility of Bitcoin during the period 2013–2019. The Halloween, day-of-the-week (DOW), and month-of-the-year (MOY) effects are analyzed. Our results reveal no evidence of a Halloween calendar anomaly. A classical DOW effect is not present in Bitcoin returns, however, we find significantly lower risk over the weekend whilst in the beginning of the week Bitcoin's volatility is more intense. Moreover, supporting evidence of a reverse January effect is detected. Our results also show that investors’ risk drops substantially in September.
This paper aims to examine the relationship between Bitcoin and preeminent financial indicators using Copula-GARCH method. In the study, we use closing prices of Bitcoin and US 10-Year Bond Yield, Gold Spot US Dollar, US Dollar Index, S&P 500, FTSE 100 and NIKKEI 225. To our knowledge, our paper is the first to examine this issue empirically. Analysis results show that there is no strong interdependence between Bitcoin and preeminent financial indicators. These findings provide new information that will benefit policy makers, banks, financial investors, and risk managers in trading activities for both long-term and short-term strategies.
This paper introduces new methods for analysing the extreme and erratic behaviour of time series to evaluate the impact of COVID-19 on cryptocurrency market dynamics. Across 51 cryptocurrencies, we examine extreme behaviour through a study of distribution extremities, and erratic behaviour through structural breaks. First, we analyse the structure of the market as a whole and observe a reduction in self-similarity as a result of COVID-19, particularly with respect to structural breaks in variance. Second, we compare and contrast these two behaviours, and identify individual anomalous cryptocurrencies. Tether (USDT) and TrueUSD (TUSD) are consistent outliers with respect to their returns, while Holo (HOT), NEXO (NEXO), Maker (MKR) and NEM (XEM) are frequently observed as anomalous with respect to both behaviours and time. Even among a market known as consistently volatile, this identifies individual cryptocurrencies that behave most irregularly in their extreme and erratic behaviour and shows these were more affected during the COVID-19 market crisis.
Jan Jakub Szczygielski, Andreas Karathanasopoulos, Adam Zaremba
We perform the most comprehensive test of cryptocurrency return distributions to date. We fit 58 hypothetical distributions to 15 major cryptocurrencies to establish which of these best describes cryptocurrency returns. The answer is: ‘It depends.’ A sharp-peaked Cauchy distribution is the most likely distribution for the majority of return series. Specific distributions are definitively identified for only a handful of cryptocurrencies. The best fitting distributions are peaked and thick-tailed, with some possessing variable shape parameters. Our findings have implications for financial modelling and its applications, such as risk measurement and risk management.
This thesis addresses the prediction problems associated with noncausal processes in cryptocurrency markets. Chapter one provides background on Bitcoin and cryptocurrencies in general. It begins by introducing four major cryptocurrencies. Then recent developments in economic research on Bitcoin are discussed. \n \nChapter two introduces a noncausal autoregressive process with Cauchy errors in application to the exchange rates of the Bitcoin electronic currency against the US Dollar. The dynamics of the daily Bitcoin/USD exchange rate series display episodes of local trends, which are modelled and interpreted as speculative bubbles. The structure of the Bitcoin market is described to give context for the presence of multiple bubbles in the exchange rate. The bubbles may result from the speculative component in the on-line trading. The Bitcoin/USD exchange rates are modelled and predicted. The mixed causal-noncausal autoregressive model is shown to better fit the data than the traditional purely causal model. A forecasting exercise using the noncausal model is then presented. \n \nChapter three examines the performance of nonlinear forecasts of noncausal processes from closed-form functional predictive density estimators. To examine the performance, time series are simulated with different conditional means and non-Gaussian distributions. The processes considered have the mixed causal-noncausal MAR(1,1)dynamics and both finite and infinite variance. The forecasts are assessed based on the forecast error behaviour and the goodness of fit of the estimated predictive density. The persistence in the noncausal component directly relates to the magnitude of the bubble effects in the time series and is found to have a meaningful impact on how forecastable the process is. To better predict bubbles the joint density of the forecast at horizon two is shown to be an effective graphical method to detect the outset of a bubble.
As crypto-currencies hold dual nature of a medium of exchange (currency) and an investment asset, some questions may arise about the potentiality of including crypto-currencies as liquid investment asset in financial institutions particularly in the banking sector to enhance their liquidity risk management and improve their portfolio diversification investment strategy. The objective of this study therefore is to examine the characteristics of Bitcoin currency based on the requirements of High-Quality Liquid Assets (HQLA) standards of Basel III and compare its volatility structure with other traditional asset classes that are already recommended by Basle III as HQLA. The study utilizes both descriptive and quantitative analysis using the GARCH family models to examine the volatility structures of these assets. The findings show that Bitcoin currency holds the same characteristics of HQLA, however; the risk of legality and recognition is still under consideration by legal authorities around the world and this risk will be eradicated in the future as crypto-currencies derive their legality from their real intrinsic value, multi-economic usefulness and not by law as in the case of fiat money currency. Furthermore, the symmetric volatility structure analysis shows the continuing persistence of volatility and predictability behavior in return series of Bitcoin currency and other- traditional asset classes in the U.S. market. However, Bitcoin’s stability has gradually improved over time. With regard to the asymmetric informative response, Bitcoin returns respond more to negative shock but it has no statistical significance, thus suggesting the lack of leveraging effect in Bitcoin market but this effect was found to be statistically persistent in other traditional asset class markets. In addition, Bitcoin returns show very low correlation with other traditional asset classes. All these imply that Bitcoin is a potential candidate as a hedge and asset diversifier, which is recommended to be included in the HQLA. This study provides some support to recent theoretical work on crypto asset return behaviour and liquidity risk management. The findings provide appropriate information about Bitcoin asset behaviour compared to other traditional asset classes which will enable them to make the right investment decision with regard to hedging, diversification and liquidity risk management. The findings of this study may assist in evaluating the suitability of including crypto assets into HQLA to improve the liquidity requirement standards and ensure that banks have an adequate amount of HQLA specifically during times of financial turmoil.
In this paper we explores as to whether cryptocurrency returns exhibit asymmetric reverting patterns and we test the presence of regime changes in the GARCH volatility dynamics of Bitcoin log-returns. For these reason, we uses non-linear autoregressive and Markov-switching GARCH (SETAR-MSGARCH) models. We finds strong evidence of regime changes in the mean and GARCH process. In addition, we conclude that bad news and good news of the same size have same impacts for investors.
Abstract One important characteristic of cryptocurrencies has been their high and erratic volatility. To represent this complicated behavior, recent studies have emphasized the use of autoregressive models frequently concluding that generalized autoregressive conditional heteroskedasticity (GARCH) models are the most adequate to overcome the limitations of conventional standard deviation estimates. Some studies have expanded this approach including jumps into the modeling. Following this line of research, and extending previous research, our study analyzes the volatility of Bitcoin employing and comparing some symmetric and asymmetric GARCH model extensions (threshold ARCH (TARCH), exponential GARCH (EGARCH), asymmetric power ARCH (APARCH), component GARCH (CGARCH), and asymmetric component GARCH (ACGARCH)), under two distributions (normal and generalized error). Additionally, because linear GARCH models can produce biased results if the series exhibit structural changes, once the conditional volatility has been modeled, we identify the best fitting GARCH model applying a Markov switching model to test whether Bitcoin volatility evolves according to two different regimes: high volatility and low volatility. The period of study includes daily series from July 16, 2010 (the earliest date available) to January 24, 2019. Findings reveal that EGARCH model under generalized error distribution provides the best fit to model Bitcoin conditional volatility. According to the Markov switching autoregressive (MS-AR) Bitcoin’s conditional volatility displays two regimes: high volatility and low volatility.
In this paper, we study forecasting problems of Bitcoin-realized volatility computed on data from the largest crypto exchange—Binance. Given the unique features of the crypto asset market, we find that conventional regression models exhibit strong model specification uncertainty. To circumvent this issue, we suggest using least squares model-averaging methods to model and forecast Bitcoin volatility. The empirical results demonstrate that least squares model-averaging methods in general outperform many other conventional regression models that ignore specification uncertainty.
Even though the empirical literature on safe haven properties of different assets with respect to financial risks is increasing, their abilities to safeguard against political risks has not been the subject of large empirical investigations. This paper uses an Empirical Mode Decomposition-based approach to look into the time-varying role of different assets (in particular, oil, precious metals and Bitcoin) as a safe haven against U.S. stocks in times of heightened uncertainty surrounding the outcome of the 2016 U.S. presidential election. Our results suggest that oil can act as an effective safe haven against political risk exposure; but such property varies over time. The abilities of gold and silver to provide positive returns during downturns have been also documented in the medium-and the long-term. Bitcoin also serves as a safe haven against U.S. stock losses but in the short-term. These findings provide useful and relevant information to investors to help ensure better asset allocation in an uncertain environment.