Laura Alessandretti, Abeer ElBahrawy, Luca Maria Aiello, Andrea Baronchelli
Machine learning and AI-assisted trading have attracted growing interest for the past few years. Here, we use this approach to test the hypothesis that the inefficiency of the cryptocurrency market can be exploited to generate abnormal profits. We analyse daily data for 1,681 cryptocurrencies for the period between Nov. 2015 and Apr. 2018. We show that simple trading strategies assisted by state-of-the-art machine learning algorithms outperform standard benchmarks. Our results show that non-trivial, but ultimately simple, algorithmic mechanisms can help anticipate the short-term evolution of the cryptocurrency market.
Stephen Chan, Jeffrey Chu, Yuanyuan Zhang, Saralees Nadarajah
In financial trading, cryptocurrencies like bitcoin use decentralization, traceability, and anonymity features to perform transactional activities. These digital currencies, using the emerging blockchain technologies, are forming the basis of the largest unregulated markets in the world. This creates various regulatory challenges, including the illicit purchase of drugs and weapons, money laundering, and funding terrorist activities. This chapter analyzes various legal and ethical implications, their effects, and various solutions to overcome the inherent issues that are currently faced by the policymakers and regulators. The authors present the result of an analysis of 30 recently published peer-reviewed scientific publications and suggest various mechanisms that can help in the detection and prevention of illegal activities that currently account for a substantial proportion of cryptocurrency trading. They suggest methods and applications that can also be used to identify the dark marketplaces in the future.
In this paper, we study the contemporaneous tail dependence structure in a pairwise comparison of the ten largest cryptocurrencies, namely Bitcoin, Dash, Dogecoin, Ethereum, Litecoin, Monero, Namecoin, Novacoin, Peercoin, and Ripple. We apply multivariate extreme value theory and we estimate a bias-corrected extreme correlation coefficient. Our findings reveal clear patterns of significantly high bivariate dependency in the distribution tails of some of the most basic and widespread cryptocurrencies, primarily over various downside constraints. This means that extreme correlation is not related to cryptocurrency market volatility per se, but to the trend of the cryptocurrency market. Therefore, extreme correlation increases in bear markets, but not in bull markets for these pairs. Interestingly, there is also a significant number of pairs which exhibit a weak level of dependency in distribution tails.
Non-linearity is characterized by an asymmetric mean-reverting property, which has been found to be inherent in the short-term return dynamics of stocks. In this paper, we explore as to whether cryptocurrency returns, as represented by Bitcoin, exhibit similar asymmetric reverting patterns for minutely, hourly, daily and weekly returns between June 2010 and February 2018. We identify several differences in the behavior of Bitcoin price returns in the pre-and post-$1,000 sub-periods and evidence of asymmetric reverting patterns in the Bitcoin price returns under all the ANAR models employed, regardless of the data frequency considered. We also present evidence indicating stronger reverting behavior of negative price returns in terms of both reverting speed and magnitude compared to positive returns and evidence of positive serial correlation with prior positive price returns. Finally, we also investigated asymmetries in Bitcoin price return series’ persistence by employing higher order ANAR models, finding evidence of a higher persistence of positive returns than negative returns, a result that further supports the existence of asymmetric reverting behavior in the Bitcoin price returns.
Trading of Bitcoin is spread about multiple venues where buying and selling is offered in various currencies. However, all markets trade one common good and by the law of one price, the different prices should not deviate in the long run. In this context we are interested in which platform is the most important one in terms of price discovery. To this end, we use a pairwise approach accounting for a potential impact of exchange rates. The contribution to price discovery is measured by Hasbrouck's and Gonzalo and Granger's information share. We then derive an ordering with respect to the importance of each market which reveals that the Chinese OKCoin platform is the leader in price discovery of Bitcoin, followed by BTC China.
We report the results of investigation of the momentum and contrarian effects on cryptocurrency markets. The investigated investment strategies involve 100 (amongst over 1200 present as of date Nov 2017) cryptocurrencies with the largest market cap and average 14-day daily volume exceeding a given threshold value. Investment portfolios are constructed using different assumptions regarding the portfolio reallocation period, width of the ranking window, the number of cryptocurrencies in the portfolio, and the percent transaction costs. The performance is benchmarked against: (1) equally weighted and (2) market-cap weighted investments in all of the ranked assets, as well as against the buy and hold strategies based on (3) S&P500 index, and (4) Bitcoin price. Our results show a clear and significant dominance of the short-term contrarian effect over both momentum effect and the benchmark portfolios. The information ratio coefficient for the contrarian strategies often exceeds two-digit values depending on the assumed reallocation period and the width of the ranking window. Additionally, we observe a significant diversification potential for all cryptocurrency portfolios with relation to the S&P500 index.
Pedro Bação, António Portugal Duarte, Hélder Sebastião, Srdjan Redžepagić
This paper investigates the information transmission between the most important cryptocurrencies -Bitcoin, Litecoin, Ripple, Ethereum and Bitcoin Cash. We use a VAR modelling approach, upon which the Geweke’s feedback measures and generalized impulse response functions are computed. This methodology allows us to fully characterize the direction, intensity and persistence of information flows between cryptocurrencies. At the availabledata granularity, most of information transmission is contemporaneous, that is, it occurs within a day. However, it seems that there are some lagged feedback effects, mainly from other cryptocurrencies to Bitcoin. The generalized impulse-response functions confirm that there is a strong contemporaneous correlation and that there is not much evidence of lagged effects. The exception appears to be related to the overreaction of Bitcoin returns to contemporaneous shocks
Purpose The purpose of this paper is to examine price overreactions in the case of the following cryptocurrencies: bitcoin, litecoin, ripple and dash. Design/methodology/approach A number of parametric ( t -test, ANOVA, regression analysis with dummy variables) and non-parametric (Mann–Whitney U -test) tests confirm the presence of price patterns after overreactions: the next day price changes in both directions are bigger than after “normal” days. A trading robot approach is then used to establish whether these statistical anomalies can be exploited to generate profits. Findings The results suggest that a strategy based on counter-movements after overreactions is not profitable, whilst one based on inertia appears to be profitable but produces outcomes not statistically different from the random ones. Therefore, the overreactions detected in the cryptocurrency market do not give rise to exploitable profit opportunities (possibly because of transaction costs) and cannot be seen as evidence against the efficient market hypothesis (EMH). Originality/value The overreactions detected in the cryptocurrency market do not give rise to exploitable profit opportunities (possibly because of transaction costs) and cannot be seen as evidence against the EMH.
Leopoldo Catania, Stefano Grassi, Francesco Ravazzolo
Cryptocurrencies have recently gained a lot of interest from investors, central banks and governments worldwide. The lack of any form of political regulation and their market far from being “efficient”, require new forms of regulation in the near future. From an econometric viewpoint, the process underlying the evolution of the cryptocurrencies’ volatility has been found to exhibit at the same time differences and similarities with other financial time-series, e.g. foreign exchanges returns. This short note focuses on predicting the conditional volatility of the four most traded cryptocurrencies: Bitcoin, Ethereum, Litecoin and Ripple. We investigate the effect of accounting for long memory in the volatility process as well as its asymmetric reaction to past values of the series to predict: 1 day, 1 and 2 weeks volatility levels.
Abstract At a given point in time, bitcoin prices are different on exchanges located in different countries, or against different currencies. While existing literature attributes the largest price differences to frictions, like market segmentation, trading platforms advertize how to execute trades based on this information. We provide a novel risk-based explanation of these price differences for a sample containing the most reputable exchanges and after accounting for all transaction costs and limitations to trade. Bitcoin prices for more expensive pairs are riskier because they depreciate more in bad times for cryptocurrency investors, when aggregate liquidity and investor sentiment are lower. (JEL G12, G14, G15, F31).
The recent evolution of cryptocurrencies has been characterized by bubble-like behavior and extreme volatility. While it is difficult to assess an intrinsic value to a specific cryptocurrency, one can employ recently proposed bubble tests that rely on recursive applications of classical unit root tests. This paper extends this approach to the case where volatility is time varying, assuming a deterministic long-run component that may take into account a decrease of unconditional volatility when the cryptocurrency matures with a higher market dissemination. Volatility also includes a stochastic short-run component to capture volatility clustering. The wild bootstrap is shown to correctly adjust the size properties of the bubble test, which retains good power properties. In an empirical application using eleven of the largest cryptocurrencies and the CRIX index, the general evidence in favor of bubbles is confirmed, but much less pronounced than under constant volatility.
Nektarios Aslanidis, Aurelio F. Bariviera, Oscar Martínez
This letter explores the behavior of conditional correlations among main cryptocurrencies, stock and bond indices, and gold, using a generalized DCC class model. From a portfolio management point of view, asset correlation is a key metric in order to construct efficient portfolios. We find that: (i) correlations among cryptocurrencies are positive, albeit varying across time; (ii) correlations with Monero are more stable across time; (iii) correlations between cryptocurrencies and traditional financial assets are negligible.
This paper aims to elucidate the connectedness between major forex currencies and cryptocurrencies using the quantile cross-spectral approach recently proposed by Baruník and Kley (2015). The sample covers six forex currencies and six cryptocurrencies over the period of 1 September 2015 to 29 December 2017. Compared with the results obtained from standard correlations and detrended moving-average cross-correlation analysis (DMCA), the quantile cross-spectral approach provides richer information on the dependence structure across different quantiles and frequencies. The most interesting result is that the intra-group dependencies are positive in the lower extreme quantiles, while inter-group dependencies are negative. This result holds in both the short- and long-term perspectives. Thus, it is worth diversifying between these two currency groups.
We use the GARCH-MIDAS model to extract the long- and short-term volatility components of cryptocurrencies. As potential drivers of Bitcoin volatility, we consider measures of volatility and risk in the US stock market as well as a measure of global economic activity. We find that S&P 500 realized volatility has a negative and highly significant effect on long-term Bitcoin volatility. The finding is atypical for volatility co-movements across financial markets. Moreover, we find that the S&P 500 volatility risk premium has a significantly positive effect on long-term Bitcoin volatility. Finally, we find a strong positive association between the Baltic dry index and long-term Bitcoin volatility. This result shows that Bitcoin volatility is closely linked to global economic activity. Overall, our findings can be used to construct improved forecasts of long-term Bitcoin volatility.
Abstract We document numerous occurrences of pump-and-dump schemes (P&Ds) targeting cryptocurrencies, which tend to trigger short-term episodes that feature dramatic increases in prices, volume, and volatility, followed by quick reversals. The evidence we document, including price run-ups before P&Ds start, suggests wealth transfers from outsiders to insiders. Our findings based on wallet-level data are consistent with the reasoning that gambling preferences, overconfidence, and naïve reinforcement learning help explain P&D participation. Finally, exploiting two natural experiments in which exchanges altered P&D policies, we find evidence consistent with the idea that P&Ds contribute to reduced cryptocurrency liquidity and lower prices.
Laura Alessandretti, Abeer ElBahrawy, Luca Maria Aiello, Andrea Baronchelli
Machine learning and AI-assisted trading have attracted growing interest for the past few years. Here, we use this approach to test the hypothesis that the inefficiency of the cryptocurrency market can be exploited to generate abnormal profits. We analyse daily data for $1,681$ cryptocurrencies for the period between Nov. 2015 and Apr. 2018. We show that simple trading strategies assisted by state-of-the-art machine learning algorithms outperform standard benchmarks. Our results show that nontrivial, but ultimately simple, algorithmic mechanisms can help anticipate the short-term evolution of the cryptocurrency market.
Abstract We establish that cryptocurrency returns are driven and can be predicted by factors that are specific to cryptocurrency markets. Cryptocurrency returns are exposed to cryptocurrency network factors but not cryptocurrency production factors. We construct the network factors to capture the user adoption of cryptocurrencies and the production factors to proxy for the costs of cryptocurrency production. Moreover, there is a strong time-series momentum effect, and proxies for investor attention strongly forecast future cryptocurrency returns.