Cryptocurrency Returns Research Articles

Has bitcoin been dethroned too quickly? The cryptocurrency return networks

Has bitcoin been dethroned too quickly? The cryptocurrency return networks

Quantile and expectile copula-based hidden Markov regression models for the analysis of the cryptocurrency market

The role of cryptocurrencies within the financial systems has been expanding rapidly in recent years among investors and institutions. It is therefore crucial to investigate this phenomenon and develop statistical methods able to capture their interrelationships, the links with other global systems and, at the same time, the serial heterogeneity. Here we introduce hidden Markov regression models for jointly estimating quantiles and expectiles of cryptocurrency returns using regime-switching copulas. The proposed approach allows us to focus on extreme returns and describe their temporal evolution by introducing time-dependent coefficients, evolving according to a latent Markov chain. Moreover to model their time-varying dependence structure, we consider elliptical copula functions defined by state-specific parameters. Maximum likelihood estimates are obtained via an expectation-maximization algorithm. The empirical analysis investigates the relationship between daily returns of five cryptocurrencies and major world market indices.

Multivariate copula-based conditional quantiles: analytic higher-order moments and ratio estimation approaches

ABSTRACT In this paper, we aim to modify multivariate copula-based conditional quantiles for a targeted random variable, given multiple random variables attaining their respective quantiles. Specifically, we propose two modifications through the Cornish–Fisher expansion, one of which involves its analytic higher-order unconditional moments. The second one accounts for its higher-order conditional moments estimated using a ratio estimation method. By considering multivariate elliptical and Archimedean copulas with Johnson's SU margins, our simulation study shows that this method provides more accurate conditional moment estimators compared to the naive method. They result in expanded conditional quantile estimators, whose efficiency is relatively better than their unexpanded versions, particularly at lower and higher quantile levels under a stronger (tail) dependence assumption. A much higher efficiency is gained when the first modification is performed. These findings are validated using an empirical study based on cryptocurrency return data and a comparative analysis against the existing estimation approaches.

Asymmetric Laplace scale mixtures for the distribution of cryptocurrency returns

AbstractRecent studies about cryptocurrency returns show that their distribution can be highly-peaked, skewed, and heavy-tailed, with a large excess kurtosis. To accommodate all these peculiarities, we propose the asymmetric Laplace scale mixture (ALSM) family of distributions. Each member of the family is obtained by dividing the scale parameter of the conditional asymmetric Laplace (AL) distribution by a convenient mixing random variable taking values on all or part of the positive real line and whose distribution depends on a parameter vector $$\varvec{\theta }$$ θ providing greater flexibility to the resulting ALSM. Advantageously concerning the AL distribution, our family members allow for a wider range of values for skewness and kurtosis. For illustrative purposes, we consider different mixing distributions; they give rise to ALSMs having a closed-form probability density function where the AL distribution is obtained as a special case under a convenient choice of $$\varvec{\theta }$$ θ . We examine some properties of our ALSMs such as hierarchical and stochastic representations and moments of practical interest. We describe an EM algorithm to obtain maximum likelihood estimates of the parameters for all the considered ALSMs. We fit these models to the returns of two cryptocurrencies, considering several classical distributions for comparison. The analysis shows how our models represent a valid alternative to the considered competitors in terms of AIC, BIC, and likelihood-ratio tests.

Intraday and daily dynamics of cryptocurrency

This paper examines and compares intraday and intraweek patterns in hourly and daily prices, returns, volumes and volatility of native cryptocurrencies, stablecoins and tokens traded on Bitstamp. We show that native cryptocurrencies and tokens share common intraday periodicity determined by the operating times of the NYSE, LSE and Hang Seng stock exchange markets. Periodic patterns are also documented in the returns on cryptocurrency market portfolio approximated by the PCA applied to intraday and intraweek cross-sectional correlation matrices of cryptocurrency returns. Stablecoins have distinct dynamics and their daily and hourly returns are uncorrelated with one another and with the returns on other cryptocurrencies. We introduce a functional CAPM to accommodate the periodic patterns and estimate it by regressing the functions of intraday and intraweek cryptocurrency returns on the market portfolio. We show that the return functions on Bitcoin, Ether, and Link satisfy affine relationships with the return functions of the market portfolio and their functional betas display periodic intraday and intraweek patterns.

Skewness risk and the cross-section of cryptocurrency returns

This paper investigates whether investors can achieve higher returns by holding cryptocurrencies with lower asymmetry risk. First, using a non-parametric bootstrap resampling method, we found that cryptocurrencies with larger market capitalizations exhibit more left-skewed performance, while those with smaller market capitalizations display more right-skewed performance. This finding is consistent with the results of Jiang et al. (2020) in the stock market. Second, both portfolio-level analyses and cross-sectional regressions at the cryptocurrency level reveal a negative cross-sectional relationship between asymmetry risk and future returns in the cryptocurrency market. Additionally, our findings indicate that skewness in the cryptocurrency market is driven by idiosyncratic risk rather than systematic risk. This contrasts with Langlois (2020), who found that systematic skewness risk outweighs idiosyncratic risk in the stock market. Finally, in addition to the risk-return tradeoff theory, the limits-to-arbitrage theory also provides explanatory insight into these results. Collectively, our findings underscore the significant role of asymmetry risk in determining cryptocurrency prices.

Application of Event Study Methodology in the Analysis of Cryptocurrency Returns

ABSTRACT This study employs event study methodology to investigate the impact of various types of events on cryptocurrency market returns and volatility. The research focuses on six major cryptocurrencies—ETH, BTC, BNB, XRP, DOGE, and TRX—over the period from December 31, 2017, to October 30, 2023. Six types of events are analyzed: cybersecurity events, block reward adjustment events, political conflict events, public health emergency events, cryptocurrency recognition and support events, and social media sentiment events. The findings reveal that cybersecurity and block reward adjustment events have minimal and short-lived impacts on market returns. Political conflict events cause significant short-term return volatility depending on market expectations. Public health emergency events, such as the COVID-19 pandemic, have significant and lasting negative impacts on market returns. Cryptocurrency recognition and support events have significant and sustained positive impacts on market returns. Social media sentiment events have significant but short-lived impacts on market returns. The robustness of the results was validated through the analysis of abnormal returns during the event period. This study provides valuable insights for investors and policymakers in managing market volatility.

A hybrid deep learning model for cryptocurrency returns forecasting: Comparison of the performance of financial markets and impact of external variables

This study introduces a finetuned hybrid forecasting model combining both Discrete Wavelet Transform (DWT) and Long Short-Term Memory network (LSTM) to predict dirty and clean cryptocurrency returns (Bitcoin and Ripple). The findings show that the proposed DWT-LSTM model outperforms a large set of benchmark models in terms of forecasting accuracy. We investigate a broader set of predictors involving financial markets (other cryptocurrencies and commodities) and external variables (blockchain information, Twitter economic uncertainty, and CO2 emissions). Our findings underline the comparable performance of the considered predictors, with the Twitter Economic Uncertainty index being the best predictor of Bitcoin returns and S&P GSCI Energy being the best predictor of Ripple returns. We also highlight the superior performance of the trading strategies based on our forecasting results.

Application and Analysis of LSTM and GRU Models for Cryptocurrency Return Prediction

Since the beginning of the 2010s, cryptocurrency has been gaining abundant attention and has become a worthy considered asset in individual investment portfolio arrangements. In this way, research on its return prediction is needed to provide guidance for investors to realize their maximum interests. This research delves into the use of prominent machine learning methods for forecasting cryptocurrency returns, with a particular emphasis on two advanced models: Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). By scrutinizing a range of academic studies, we identify the strengths and weaknesses of these models in return prediction and offer a comparative analysis. Our findings reveal that the GRU model excels by achieving lower values in both Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE), highlighting its superior predictive accuracy. Meanwhile, LSTM presents plausible recall rates, precision, accuracy, and lower cross-entropy losses. From this research, individual investors and financial organizations can learn the applications of advanced machine learning algorithms in predicting cryptocurrency return and making informed investment decisions. By comparison analysis, this paper also indicates future directions of improving machine learning algorithms to be more adjustable to the cryptocurrency environment, improving prediction accuracy.

Statistical modelling of cryptocurrencies

There has been tremendous interest invested by researchers and academics in Bitcoin since it's introduction to the financial market. However, in recent years there has been an advancement of the cryptocurrency market where other cryptocurrencies such as Ethereum, Litecoin and Ripple have grown relatively quickly and could potentially challenge the dominant placement of Bitcoin. These cryprocurrencies have been utilized globally as a virtual currency for multiple transactions. The returns of cryptocurrencies are known to be volatile and have been observed to fluctuate quite a bit in recent times. This study assesses and differentiates the performance of generalized autoregressive score (GAS) models integrated with a few heavy-tailed distributions in Value-at-Risk (VaR) estimation of the four most popular cryptocurrencies' returns, i.e. Bitcoin returns, Ethereum returns, Litecoin returns and Ripple returns. This paper proposed VaR models for Bitcoin, Ethereum, Litecoin and Ripple returns, i.e. GAS models combined with the generalized hyperbolic distribution (GHD), the variance gamma (VG) distribution, the normal inverse Gaussian (NIG) distribution and the generalized lambda distribution (GLD). The Kupiec likelihood ratio test was adopted to evaluate the proposed models' adequacy and Backtesting VaR was used to select the superior set of models.

Beyond Automation: Unpacking the Role of Trading Bots in Shaping Online Cryptocurrency Return

Beyond Automation: Unpacking the Role of Trading Bots in Shaping Online Cryptocurrency Return

Forecasting the Risk of Cryptocurrencies: Comparison and Combination of GARCH and Stochastic Volatility Models

Abstract The high returns of cryptocurrencies have attracted many investors in recent years. At the same time the evolution of cryptocurrencies is characterized by extreme volatility. For investors, it is therefore key to gauge the risks related to an investment in cryptocurrencies. We provide a comparison of several GARCH and stochastic volatility models for forecasting the risk of cryptocurrencies over the out-of-sample period from 28.09.2018 to 28.02.2023. It turns out that the widely used GARCH(1,1) does not provide accurate risk predictions. In contrast, adding t-distributed innovations or allowing for regime changes improves the accuracy in both model classes. Finally, we consider a Bayesian decision-guided approach with discount learning to combine the different models and provide robust evidence that combining the model predictions leads to accurate combined risk predictions.

The Impact of the Cryptocurrency Market on Islamic vs. Conventional Stock Returns: Evidence from Gulf Cooperation Council Countries

The rapid rise and widespread global adoption of cryptocurrencies in recent years has fundamentally transformed the international financial landscape, with digital assets increasingly being recognized for their potential to influence the stability and performance of traditional capital markets. Against this backdrop, this study aims to empirically investigate the impact of cryptocurrency returns on Islamic vs. conventional stock returns in Gulf Cooperation Council (GCC) countries. The salient distinctions between Islamic and conventional stock markets include fundamental differences in principles, investment allocations, and risk profiles, underscoring the importance of examining the impact of cryptocurrency returns on these distinct equity segments. Daily data were collected from stock indices in five GCC countries over the period 2016–2019, including two sub-periods: before and after the 2017 crypto crash. Pooled OLS, fixed effects, random effects, and generalized linear models (GLMs) were used to analyze the data collected during the study. With the GCC increasingly focusing on cryptocurrency markets, there is growing concern about these markets’ potential impact on regional stocks. This study addresses the important questions of whether the impacts of the cryptocurrency market on Islamic vs. conventional stock markets differ throughout the GCC region and how these impacts have evolved since the crypto crash period. The findings reveal that cryptocurrency returns had a negative impact on both GCC Islamic and conventional stock market returns for the full sample period (2016–2019), and the negative effect was far more pronounced for conventional stocks. For the two sub-periods before and after the crash, only the cryptocurrency market and conventional GCC stocks remained negatively correlated, while the cryptocurrency market and the GCC Islamic stock markets became uncorrelated. Thus, for the calmer sub-periods before and after the crypto crash, the rise in cryptocurrency returns may have enticed GCC investors away from conventional stocks, perhaps resulting in a decline in their investment in these stocks. Meanwhile, those who invest in Islamic stocks may not be exposed to this temptation.

Cryptocurrency returns and consumption-based asset pricing

ABSTRACT This paper examines whether the cross-section of cryptocurrency returns is captured by risk factors based on consumption-based asset pricing. It is an imperative task for financial economists to find the fundamental risk behind characteristic-based cryptocurrency factors in order to economically understand cryptocurrency. To address the data availability issue in the analysis of cryptocurrency, we employ mixed data sampling (MIDAS) regression to check the relation between the principal component analysis (PCA) factors in cryptocurrency returns and the factors in the consumption capital asset pricing models (CCAPMs) and intertemporal capital asset pricing models (ICAPMs). We establish significant links between them, which in turn implies that cryptocurrency returns are investors’ compensation for bearing consumption risk, conditional consumption risk, and intertemporal consumption risk. This finding underscores that cryptocurrency returns are the manifestation of macroeconomic equilibrium derived from investors’ utility maximization.

Support vector regression-based heteroscedastic models for cryptocurrency risk forecasting

In this study, we aimed to combine support vector regression (SVR) with an autoregressive (AR) model and a generalized autoregressive conditional heteroscedastic (GARCH) or asymmetric GARCH model. We compared the performances of these combined models with the maximum likelihood estimation (MLE)-based models under a normal, Student’s t, or skewed Student’s t distributional assumption. We employed such models to protect the heteroscedastic risks of four major cryptocurrencies, namely, Bitcoin, Ethereum, Tether, and Binance Coin, before and during COVID-19. We accomplished this task by forecasting their volatility, Value-at-Risk (VaR), and Expected Shortfall (ES). The empirical results revealed that in terms of the root-mean-square error (RMSE) and mean-absolute error (MAE) metrics, the SVR-based models with a polynomial or radial basis kernel function, which is nonlinear, resulted in a more accurate volatility forecast compared to the MLE-based models under any distributional assumption. The superiority of the former models was more apparent when applied to Tether exhibiting the most stable volatility. This supports evidence of nonlinear characteristics in the cryptocurrency return and volatility series. A combination between SVR and an asymmetric GARCH model also confirms the presence of an asymmetrically nonlinear relationship between the returns and volatility. In addition, the correct-VaR and Kupiec’s backtesting results showed that the SVR-based models tended to produce the VaR forecasts for the four cryptocurrencies with better accuracy at the 1%, 5%, and 10% significance levels compared to the MLE-based ones. The use of the correct-ES metric also leads to the superior ES forecasting performance of the former models at any level. However, McNeil and Frey’s backtesting results demonstrated their worse ES forecasting performance at the 1% level. As the COVID-19 pandemic progressed, we also found an increase in the Bitcoin and Binance Coin market risks as well as a decrease in the Ethereum and Tether market risks.

Forecasting cryptocurrency returns using classical statistical and deep learning techniques

The emergence of cryptocurrencies has generated enthusiasm and concern in the modern global economy. However, their high volatility, erratic price fluctuations, and tendency to exhibit price bubbles have made investors cautious about investing in them. Consequently, it is essential to develop methods and models to forecast cryptocurrency returns to benefit investors, traders, and the scientific community. Despite the considerable volume of research on Bitcoin price forecasting, other cryptocurrencies have received little attention in academic literature. Additionally, the current body of literature on predicting cryptocurrency prices or returns emphasizes the use of in-sample methodologies. However, this method is susceptible to overfitting. To address these gaps in the literature, this study employs autoregressive moving average (ARMA), generalized autoregressive conditional heteroskedasticity (GARCH), exponential generalized autoregressive conditional heteroskedasticity (EGARCH), and long short-term memory (LSTM) deep learning neural networks to forecast returns for the ten most actively traded digital currencies: Bitcoin, Ethereum, Ripple, Chainlink, Litecoin, Cardano, Ethereum Classic, Bitcoin Cash, Tether, and Binance Coin. To assess the accuracy of the two models, this study utilizes an out-of-sample method with data gathered sequentially from November 9, 2017, to September 18, 2022. The results indicate that all models exhibit high accuracy, as evidenced by their low root mean square error (RMSE), mean absolute error (MAE), and mean squared error (MSE) values. Meanwhile, the hybrid EGARCH-LSTM or GARCH-LSTM models demonstrate slightly better accuracy compared with the other models. The findings are valuable for investors, traders, and researchers involved in cryptocurrency forecasting.

Price Delay and Market Efficiency of Cryptocurrencies: The Impact of Liquidity and Volatility during the COVID-19 Pandemic

The rise of cryptocurrencies as alternative financial investments, with potential safe-haven and hedging properties, highlights the need to examine their market efficiency. This study is the first to investigate the combined impact of liquidity and volatility features of cryptocurrencies on their price delays. Using a wide spectrum of cryptocurrencies, we investigate whether the COVID-19 outbreak has affected market efficiency by studying price delays to market information. We find that as liquidity increases and volatility decreases, cryptocurrencies demonstrate stronger market efficiency. Additionally, we show that price delay differences during the COVID-19 outbreak increase with higher levels of illiquidity, particularly for highly volatile quintiles. We suggest that perceived risks and high transaction costs in illiquid and highly volatile cryptocurrencies reduce active traders’ willingness to engage in arbitrage trading, leading to increased market inefficiencies. Our findings are relevant to investors, aiding in improving their decision-making processes and enhancing their investment efficiency. Our paper also presents significant implications for policymakers, emphasizing the need for reforms aimed at enhancing the speed at which information is incorporated into cryptocurrency returns. These reforms would help mitigate market distortions and increase the sustainability of cryptocurrency markets.

Intricacy of cryptocurrency returns

This paper quantifies the intricacy, i.e., non-linearity and interactions of predictor variables, in explaining cryptocurrency returns. Using data from several thousand cryptocurrencies spanning 2014 to 2022, we observe a notably high level of intricacy. This provides a quantitative measure why linear models are often outperformed by machine learning algorithms in predicting cryptocurrency returns. Furthermore, we document that the intricacy in these predictions is considerably larger compared to stocks. Our analysis reveals that interactions are gaining importance over time, while individual non-linearity of the drivers is diminishing. This adds to the emerging literature on spillover effects between cryptocurrencies, traditional finance and the economy. This finding is important for investors as well as regulators as the high intricacy proposes challenges to both actors in the market.

Optimizing Cryptocurrency Returns: A Quantitative Study on Factor-Based Investing

This study explores cryptocurrency investment strategies by adapting the robust framework of factor investing, traditionally applied in equity markets, to the distinctive landscape of cryptocurrency assets. It conducts an in-depth examination of 31 prominent cryptocurrencies from December 2017 to December 2023, employing the Fama–MacBeth regression method and portfolio regressions to assess the predictive capabilities of market, size, value, and momentum factors, adjusted for the unique characteristics of the cryptocurrency market. These characteristics include high volatility and continuous trading, which differ markedly from those of traditional financial markets. To address the challenges posed by the perpetual operation of cryptocurrency trading, this study introduces an innovative rebalancing strategy that involves weekly adjustments to accommodate the market’s constant fluctuations. Additionally, to mitigate issues like autocorrelation and heteroskedasticity in financial time series data, this research applies the Newey–West standard error approach, enhancing the robustness of regression analyses. The empirical results highlight the significant predictive power of momentum and value factors in forecasting cryptocurrency returns, underscoring the importance of tailoring conventional investment frameworks to the cryptocurrency context. This study not only investigates the applicability of factor investing in the rapidly evolving cryptocurrency market, but also enriches the financial literature by demonstrating the effectiveness of combining Fama–MacBeth cross-sectional analysis with portfolio regressions, supported by Newey–West standard errors, in mastering the complexities of digital asset investments.

Revisiting seasonality in cryptocurrencies

The evidence on seasonality in cryptocurrency returns is not robust. Although the positive Monday effect for Bitcoin is internally valid, it does not persist in data after 2015. We do not find robust evidence of return abnormalies but of lower trading activity on weekends. This finding is robust across 500 different coins. We also find that the Monday effect in the cross-section of coins is typically negative, but the confidence intervals remain wide.