The machine learning method has been used in stock price prediction for a long time, and the price of cryptocurrencies such as bitcoin has attracted more and more attention in recent years. This paper aims to improve the method applicable to the stock market and try to use it in cryptocurrency price prediction. A simple three-layered feedforward artificial neural networks (ANN) model was applied in this paper to predict the daily directions of cryptocurrency prices. The historical trading data of Bitcoin, Ethereum, and Cardano were used in the experiments. Nine selected technical indicators were preprocessed into discrete trend data, and they were input into the model together with three additional indicators for training. This study has preliminarily obtained an effective result with price prediction accuracy of the three cryptocurrencies between 61% and 65%.
Purpose The purpose of this study is to compare five data-driven-based ML techniques to predict the time series data of Bitcoin returns, namely, alternating model tree, random forest (RF), multiple linear regression, multi-layer perceptron regression and M5 Tree algorithms. Design/methodology/approach The data used to forecast time series data of Bitcoin returns ranges from 8 July 2010 to 30 Aug 2020. This study used several predictors to predict bitcoin returns including economic policy uncertainty, equity market volatility index, S&P returns, USD/EURO exchange rates, oil and gold prices, volatilities and returns. Five statistical indexes, namely, correlation coefficient, mean absolute error, root mean square error, relative absolute error and root relative squared error are determined. The results of these metrices are used to develop colour intensity ranking. Findings Among the machine learning (ML) techniques used in this study, RF models has shown superior predictive ability for estimating the Bitcoin returns. Originality/value This study is first of its kind to use and compare ML models in the prediction of Bitcoins. More studies can be carried out by using further cryptocurrencies and other ML data-driven models in future.
This paper investigates the relationship between the COVID-19 crisis and the two leading cryptocurrencies, Bitcoin and Ethereum, from 31 December 2019 to 18 August 2020. We also use an economic news sentiment index and financial market sentiment index to explore the possible mechanisms through which COVID-19 impacts cryptocurrency. We employ a VAR Granger Causality framework and Wavelet Coherence Analysis and find the cryptocurrency market was impacted in the early phase of the sample period through economic news and financial market sentiments, but this effect diminished after June 2020.
Bitcoins are evolving as a modern class of investment assets and it is crucial for investors to manage their investment risk. This paper examines the impact of macroeconomic-financial indicators on Bitcoin price using symmetric and asymmetric version of autoregressive distributed lag (ARDL) models with structural breaks. The asymmetric long-run association ascertained between Bitcoin prices and the macroeconomic-financial indicators is evident. Our empirical results indicate that the Bitcoin cannot be used to hedge against the inflation, Federal funds rate, stock markets and commodity markets. We further find that Bitcoin can be regarded as a hedging device for the oil prices. Our findings have significant implications for market participants who consider including alternate investment assets in their portfolios.
This primer provides a pairwise comparison of cryptocurrency characteristics with those of fiat currency and hard commodities to shed light on the nexus between cryptocurrencies, fiat currency, and hard commodities. Then, it synthesizes methods and results from empirical research that investigate the nexus. The findings reveal that the existing literature has not reached a consensus on the nature of cryptocurrencies, in particular whether they should be categorized as a currency or a commodity which indicates the research area is not yet saturated.
Sonia Arsi, Soumaya Ben Khelifa, Yosra Ghabri, Héla Mzoughi
Cryptocurrencies are witnessing a growing interest from investors and the media. They are increasingly perceived as a new class of assets through added benefits, such as hedging capabilities and diversification. However, this does not preclude the fact that cryptocurrencies can be risky assets. Within such a context, diverse studies were carried out at various risk levels. Our chapter bridges this gap and tries to reconcile varying positions on risk across cryptocurrencies. Particularly, we provide a detailed overview on the main risks to be considered by crypto-traders, namely, technology, fraud, legal, market, liquidity, and COVID-19 pandemic risks. The main findings show that the occurrence of any technological failure tends to raise insecurity and distrust in the cryptocurrency technology. This fact can be even further spoiled through fraud schemes and fake trading volumes. Additionally, the legal framework of the cryptocurrencies is still inconclusive. In terms of market risk, these crypto-assets are riskier than fiat currencies and there is a significant risk contagion across large-cap cryptocurrencies. Then, a significant relationship exists between liquidity and efficiency in the cryptocurrency market, since price dynamics can influence the market liquidity. Finally, it sorts out that the COVID-19 pandemic heavily affected the cryptocurrency markets. This chapter underlines current challenges for investors, regulators, and policymakers.
The cryptocurrency market debate resumed in 2020 with renewed vigour as the price of Bitcoin surpassed late 2017 highs. This study aims to analyse possible factors of Bitcoin’s pricing at various cryptocurrency market development stages — before the 2017 price bubble, after and during the COVID-19 pandemic. The main method of analysis is a generalized autoregressive conditional heteroskedasticity model with conditional generalized error distribution (GARCHGED). Two groups of indicators are used as possible factors related to the Bitcoin dynamics. The first group consists of various quantitative indicators directly related to Bitcoin (the so-called internal factors) — the volume of exchange trade, the volume of transactions in the Bitcoin blockchain, the number of new and active wallets, hash rate, the sum of fees paid in the blockchain, as well as the dynamics of Google Trends search queries. The second group is the return on various financial assets — stock and bond indexes, commodities, and currency markets. The results of the analysis demonstrate the absence of a stable correlation between any of the factors under consideration and Bitcoin returns in all the periods that we focus on. In the period before the 2017 price bubble, the internal factors and Bitcoin returns showed generally co-directional dynamics, but the situation changed in 2018. In early 2021, the correlation between Bitcoin and traditional financial assets returns has increased significantly. We can conclude that Bitcoin is becoming a popular means of diversification as a high-risk asset, which, however, follows the pattern of a speculative bubble at the beginning of 2021. The increased demand for the need to invest in Bitcoin using various exchange-traded instruments (ETFs for cryptocurrencies) may soon lead to a further increase in the price of this cryptocurrency if such instruments are registered on the exchange.
The pandemic of coronavirus (COVID-19) creates fear and uncertainty causing extraordinary disruption to financial markets and global economy. Witnessing the fastest selloff in the American stock market in history with a plunge of more than 28% in S&P 500 has increased the volatility of global financial market to exceed the level observed during the financial crisis of 2008. On the other hand, Bitcoin value has shown considerable stability in the last couple of months peaking at $10,367.53 in the mid of February 2020. In this context, the aim of this paper is to investigate the impact of COVID-19 numbers on Bitcoin price taking into consideration number of controlling variables including WTI-oil price, S&P 500 index, financial market volatility, gold prices, and economic policy uncertainty of the US. To do so, ARDL estimation has been applied using daily data from December 31, 2019 till May 20, 2020. Key findings reveal that the daily reported cases of new infections have a marginal positive impact on Bitcoin price in the long term. However, the indirect impact associated with the fear of COVID-19 pandemic via financial market stress cannot be neglected. Bitcoin can also serve as a hedging tool against the economic policy uncertainty in the long term. In the short run, while the returns of economic policy uncertainty have no impact on Bitcoin price, the growth in the new cases of COVID-19 infection and returns of financial market volatility have more positive significant impact on Bitcoin returns.
Due to the nonlinearity and highly volatile dynamics of the price data of cryptocurrency, classic parametric models show limited success in tracking and prediction. With the rise of deep learning recently, various researches on forecasting the price of cryptocurrency using deep neural network have reported encouraging results in the cases of low volatility. In this study, we propose a hybrid approach which combines the advantages of non-stationary parametric models such as Generalized Autoregressive Conditional Heteroskedasticity (GARCH) with the nonlinear modelling potential of Long-Short Term Memory (LSTM) neural networks. The results show that our hybrid model has a similar predictive performance in terms of MSE, MAE and RMSE but higher metric scores in precision, accuracy and F1 score under optimal hyperparameters. This study reveals that the combination of parametric models like GARCH with deep neural network may come up with better results in cryptocurrency price forecasting especially in the case of highly volatile data or when short data sequences are available. Moreover, the proposed framework can be used also in other applications where high volatility and scarcity of data are the main characteristics.
Aastha Agarwal, S Keerthana, Rahul Reddy, Afraz Moqueem
Bitcoin, Etherium, and Litecoin are among the most extensive market capitalized cryptocurrencies in the present era. With the increased popularity, there is also an increased proclivity of investors towards investing in cryptocurrencies. To gain maximum profits and avoid risks, one needs to analyze the trends and history of the cryptocurrency diligently. This paper put forth various machine learning algorithms to scrutinize cryptocurrencies such as Bitcoin, Etherium, and Litecoin based on multiple trading factors such as open price, close price, volume, market price, history, etc. We have performed various state-of-the-art machine learning to predict the future market value of the cryptocurrencies and derived the performance analysis of the same.
Marion Labouré, Markus Müller, Gerit Heinz, Sagar Singh · 5 authors
Abstract This paper aims to provide an outline of the dynamic landscape of cryptocurrencies and central bank digital currencies (CBDC) so as to comment about the role and prospect of both in the future. We highlight the main drivers of the ongoing digital currency wave from a socio‐economic as well as historical perspective. From an investment standpoint, we evaluate the merits of placing a cryptocurrency within a diversified portfolio and analyse other factors to be taken into account when it comes to asset allocation considerations. We also explore environmental, social, and governance (ESG) implications of the introduction of such digital currencies. Finally, we comment on the current status of national CBDC projects and what it would mean for the digital currency universe when the official CBDC roll‐outs begin in a few years.
Barbara Będowska-Sójka, Agata Kliber, Aleksandra Rutkowska
We try to establish the commonalities and leadership in the cryptocurrency markets by examining the mutual information and lead-lag relationships between Bitcoin and other cryptocurrencies from January 2019 to June 2021. We examine the transfer entropy between volatility and liquidity of seven highly capitalized cryptocurrencies in order to determine the potential direction of information flow. We find that cryptocurrencies are strongly interrelated in returns and volatility but less in liquidity. We show that smaller and younger cryptocurrencies (such as Ripple's XRP or Litecoin) have started to affect the returns of Bitcoin since the beginning of the pandemic. Regarding liquidity, the results of the dynamic time warping algorithm also suggest that the position of Monero has increased. Those outcomes suggest the gradual increase in the role of privacy-oriented cryptocurrencies.
This paper establishes a brand-new perspective of analyzing the risk of crypto assets through a semi-nonparametric approach, discussing its theoretical advantages and testing its performance compared to parametric approaches and in terms of backtesting techniques and different risk measures: Value-at-Risk, Expected Shortfall and Median Shortfall. Our comprehensive analysis for six cryptocurrencies shows that flexible semi-nonparametric approaches outperform risk measures of most crypto assets (particularly Bitcoin) and tend to provide the most conservative risk assessment. Furthermore, we propose the Median Shortfall as a robust-to-outliers and reliable risk measure for cryptocurrencies and discuss on the choice of the appropriate probability levels according to the assumed distribution. The evidence supports that Median Shortfall at 98.31 % and 98.51 % confidence levels as accurate alternatives to Value-at-Risk at 99 % and Expected Shortfall at 97.5 %.
Abstract This chapter examines the dynamic linkages between the returns of Bitcoin, gold, and oil by using daily closing price data between July 17, 2010 and January 8, 2021. This study applies the diagonal BEKK–GARCH model for the purpose of analyzing a volatility spillover of variables in positive or negative ways. The empirical results show that the lagged returns inversely affect their current returns in oil. Based on the return spillovers between Bitcoin and gold, the empirical results indicate a unidirectional return spillover from Bitcoin to gold. Moreover, the authors found a unidirectional return transmission is observed from oil to Bitcoin, implying that oil returns are useful in forecasting Bitcoin returns. These findings are not only valuable for understanding of the interrelationships between the returns of Bitcoin, gold, and oil, but they are also of great interest to portfolio managers, investors, and investment funds that are actively dealing in Bitcoin, gold, and oil.
Sinan Erdoğan, Maruf Yakubu Ahmed, Samuel Asumadu Sarkodie
Abstract When Bitcoin (BTC), the first pioneering cryptocurrency was released in 2009, it was considered as an apolitical currency. Besides, the possible effect of BTC and other cryptocurrencies on either financial markets or transactions has been widely discussed. However, the environmental effects of cryptocurrency demand have been ignored. Here, this study examines the nexus between cryptocurrencies and environmental degradation by employing standard and asymmetric causality methods. The Toda-Yamamoto and bootstrap-augmented Toda-Yamamoto test results reveal Bitcoin and Ethereum (ETH) excluding Ripple (XRP) have causal effects on environmental degradation. The Fourier-augmented Toda-Yamamoto test results show causal effects running from Bitcoin and Ripple to environmental degradation, whereas no causal effect runs from Ethereum to environmental degradation. The asymmetric causality shows causal effects from the positive shock of Bitcoin demand, negative shocks of Ripple and Ethereum demands to positive shocks of environmental degradation. Further discussions and policy implications are provided in the relevant sections of this study.
Risk management and prediction of market losses of cryptocurrencies are of notable value to risk managers, portfolio managers, financial market researchers and academics. One of the most common measures of an asset’s risk is Value-at-Risk (VaR). This paper evaluates and compares the performance of generalized autoregressive score (GAS) combined with heavy-tailed distributions, in estimating the VaR of two well-known cryptocurrencies’ returns, namely Bitcoin returns and Ethereum returns. In this paper, we proposed a VaR model for Bitcoin and Ethereum returns, namely the GAS model combined with the generalized lambda distribution (GLD), referred to as the GAS-GLD model. The relative performance of the GAS-GLD models was compared to the models proposed by Troster et al. (2018), in other words, GAS models combined with asymmetric Laplace distribution (ALD), the asymmetric Student’s t-distribution (AST) and the skew Student’s t-distribution (SSTD). The Kupiec likelihood ratio test was used to assess the adequacy of the proposed models. The principal findings suggest that the GAS models with heavy-tailed innovation distributions are, in fact, appropriate for modelling cryptocurrency returns, with the GAS-GLD being the most adequate for the Bitcoin returns at various VaR levels, and both GAS-SSTD, GAS-ALD and GAS-GLD models being the most appropriate for the Ethereum returns at the VaR levels used in this study.
Yuzhi Cai, Thanaset Chevapatrakul, Danilo V. Mascia
Abstract We shed light on how the price explosivity characterising Bitcoin and other major cryptocurrencies is triggered, by employing the Quantile Self-Exciting Threshold Autoregressive (QSETAR) model. Our results for Bitcoin, Ripple, and Stellar reveal that the explosive behaviour originates from the extreme upper tails of the return distributions following a price increase in the preceding day. We do not find evidence of explositivity in the price of Litecoin.
The cryptocurrency market has experienced stunning growth, with market value exceeding USD 1.5 trillion. We use a DCC-MGARCH model to examine the return and volatility spillovers across three distinct classes of cryptocurrencies: coins, tokens, and stablecoins. Our results demonstrate that conditional correlations are time-varying, peaking during the COVID-19 pandemic sell-off of March 2020, and that both ARCH and GARCH effects play an important role in determining conditional volatility among cryptocurrencies. We find a bi-directional relationship for returns and long-term (GARCH) spillovers between BTC and ETH, but only a unidirectional short-term (ARCH) spillover effect from BTC to ETH. We also find spillovers from BTC and ETH to USDT, but no influence running in the other direction. Our results suggest that USDT does not currently play an important role in volatility transmission across cryptocurrency markets. We also demonstrate applications of our results to hedging and optimal portfolio construction.
Machine learning (ML) algorithms have been widely used to predict future financial trends. It has become a tool for predicting future trends based on what is known beforehand. Like other financial stock markets, cryptocurrency has become a new sensation and challenge for investors to predict its behaviour. However, unlike other financial instruments, cryptocurrency has been renowned because of the difficulty to predict the price due to its volatility behaviour that changes so rapidly and since there is no fundamental economy for its value. This paper presents a performance comparison of two ML algorithms in predicting Ethereum price with non-time series analysis, which are k- Nearest Neighbors (k-NN) and multiple polynomial regression (MPR). The experiment used independent variables from related real-world economic fundamentals such as Dow Jones Index, gold price, oil price, and Ethereum volume. The experiment data was collected from the records from April 2017 until April 2021. For each algorithm, several methods of preprocessing data were used to match all independent data with the dependent data. Three different preprocessing scenarios were also used to find the maximum accuracy model. scenario 1 (feature selection based on correlation matrix), scenario 2 (feature selection based on correlation with the dependent variables and among independent variables), and scenario 3 (scenario 1 extracted with PCA). The performance of the compared methods was evaluated by using MSE and MAE. From the experiment, a comparison of results using two different models with k-NN and multiple polynomial regression is obtained. It is found that k-NN with a hyperparameter K = 2 have the best prediction with MSE = 449.032 and MAE = 14.282 compared with multiple polynomial regression with the best MSE = 13953.96 and MAE = 84.923.
This paper studied the mean and volatility transmission among Bitcoin as the most prominent cryptocurrency, exchange rates from developed countries/regions, and exchange rates from emerging countries/regions. Using daily returns between January 1, 2015, and December 31, 2018, and Bivariate VAR - Diagonal VECH models. The empirical results suggest there was no mean transmission between USD/EUR and USD/BTC. However, there was a unidirectional mean shock transmission link from USD/CNH, USD/MAD, and USD/IDR to USD/BTC. The results also suggested the existence of a bidirectional cross-volatility persistence link between bitcoin and all the exchange rates, except for USD/IDR and a bidirectional cross-volatility spillover link between USD/BTC and USD/CNH. A critical implication of these results is that they will be of use to investors, speculators, risk managers, and policymakers in understanding the degree of integration in terms of volatility and return among Bitcoin, currencies from developed, and currencies from emerging countries.