Cryptocurrency Time Series Research Articles

CARROT: Simultaneous prediction of anomalies from groups of correlated cryptocurrency trends

Cryptocurrencies are virtual currencies that exploit cryptography to perform secure financial transactions. They gained widespread popularity in recent years due to their decentralized nature, (pseudo-)anonymity, and ability to facilitate cross-border transactions without the need for intermediaries. However, their price on the market exhibits a huge volatility, that makes them prone to market anomalies. Therefore, predicting anomalies in cryptocurrency time series can be considered an important task for financial institutions, traders, and investors, to maximize their profit or minimize losses.In this paper, we propose a novel approach for predicting anomalies in cryptocurrency time series by exploiting temporal correlations among different cryptocurrencies. Our approach, called CARROT, is based on the idea that groups of cryptocurrencies exhibit similar trends, possibly due to common influencing factors. CARROT analyzes the temporal correlation between different cryptocurrencies, and identifies clusters showing similar patterns that can be useful for gaining insights into future anomalies. Subsequently, CARROT exploits multiple (i.e., one for each cluster) multi-target LSTM models to predict anomalies.Our experiments, performed on a dataset of 17 cryptocurrencies, proved that CARROT outperforms single-target LSTM models of up to 20%, as well as other approaches based on neural networks, i.e., MLP and CNN, in terms of macro F1-score. Therefore, the proposed approach can be considered as a promising tool for predicting anomalies in cryptocurrency time series data and can potentially be used to improve risk management and trading strategies in the cryptocurrency market.

Forecasting and Analysis of Cryptocurrency Trends

The development of financial technology has given rise to a new class of asset known as cryptocurrency, which has presented a significant potential for study. Forecasting cryptocurrency prices is challenging because of their dynamic and volatile nature. There are hundreds of different cryptocurrencies in circulation today. With prospects in peer-to-peer payments, remittance, e-commerce and retail, and media & entertainment, the cryptocurrency business appears to have a bright future. From 2019 to2024, the worldwide cryptocurrency industry is anticipated to expand at a 32% CAGR. The distributed ledger technology's immutability and transparency, the rise in remittances from poor nations, shifting monetary laws, and a sharp rise in venture capital investments are the main factors propelling this market's growth. We created a short-term prediction (between five and thirty days) for the cryptocurrencies using the ARIMA model. When it came to predicting cryptocurrency time series, we discovered that the suggested method outperformed the ARIMA-ARFIMA models in both periods of gradual rising (falling) and transition dynamics (change of trend). Index Terms: digital currency,cryptocurrency, and ARIMA model.

Spillover effects, lead and lag relationships, and stable coins time series

This research explores plausible spillover effects among S&P 500 Index, stable coins, and selected cryptocurrency time series by examining observable lead and lag interrelationships among the series. Considering the heteroscedastic and “fat-tailed” nature of the data distributions underlying the empirical analyses, we employ quantile Granger Causality tests to improve the validity of our statistical findings. Our test results suggest that stable coins, USDT, and USDC offer diversification benefits by decreasing portfolio risk. The log returns of the S&P 500 Index, stable coins, Bitcoin, Ethereum, and Binance coins demonstrate clear bidirectional causality and spillover effects in low and high quantiles. Interestingly, however, stable coin and USDT returns strongly lead S&P 500 Index returns in nearly all quantiles for post COVID-19 time periods. These findings indirectly support intuition based upon market co-movement or integration assertions and suggest that investors can obtain added diversification benefits deriving from causality or spillover effects of holding stable coins when forming investment portfolios.

A Novel Feature Selection Method for Risk Management in High-Dimensional Time Series of Cryptocurrency Market

In this study, a novel approach for feature selection has been presented in order to overcome the challenge of classifying positive and negative risk prediction in the cryptocurrency market, which contains high fluctuation. This approach is based on maximizing information gain with simultaneously minimizing the similarity of selected features to achieve a proper feature set for improving classification accuracy. The proposed method was compared with other feature selection techniques, such as sequential and bidirectional feature selection, univariate feature selection, and least absolute shrinkage and selection operator. To evaluate the feature selection techniques, several classifiers were employed: XGBoost, k-nearest neighbor, support vector machine, random forest, logistic regression, long short-term memory, and deep neural networks. The features were elicited from the time series of Bitcoin, Binance, and Ethereum cryptocurrencies. The results of applying the selected features to different classifiers indicated that XGBoost and random forest provided better results on the time series datasets. Furthermore, the proposed feature selection method achieved the best results on two (out of three) cryptocurrencies. The accuracy in the best state varied between 55% to 68% for different time series. It is worth mentioning that preprocessed features were used in this research, meaning that raw data (candle data) were used to derive efficient features that can explain the problem and help the classifiers in predicting the labels.

Investment strategies based on anomalies detected in the financial time series of cryptocurrencies

The purpose of the research is to study the cryptocurrency data listed on Binance, and design a profitable strategy based on the findings. The data covers over 150 selected cryptocurrencies. The study aims to detect anomalies in the volume and number of transactions and apply an investment strategy based on deviations and sudden price fluctuations. An autoencoder and LSTM-based neural network have been used. Based on the results of the present research, it can be concluded that the model successfully identified anomalies in the data regarding the volume and number of transactions carried out. I it was also observed that price volatility in the period close to the detected anomaly was significantly higher than average volatility for the sample.

Investigating long and short memory in cryptocurrency time series by stochastic fractional Brownian models

Over the last years, the world of cryptocurrencies has undergone a tumultuous development, mostly characterized by speculative behaviors, and thus one might argue that it does not satisfy the Efficient Market Hypothesis. Since the efficiency of a financial market can be assessed by checking whether the times series of the assets traded on it are persistent/anti-persistent, we investigate the presence of memory in the price of seven among the most important cryptocurrencies. To this aim, we employ an original approach based on two fractional models, namely the geometric fractional Brownian motion and the geometric mixed fractional Brownian motion, which are tested against the Markovian geometric Brownian motion to detect the presence of memory. The above fractional models are estimated by employing an innovative maximum likelihood procedure that exploits the Toeplitz structure of the covariance matrix of log-returns. The null assumption of absence of memory in the time series is tested based on confidence ellipses for the estimated parameters. We validate the proposed procedure on artificial data and we apply it to the historical series of crypto-prices. Results suggest the presence of memory effects only for some of the considered cryptocurrencies. A possible explanation of such a persistence/anti-persistence phenomenon is provided.

A novel hybrid walk-forward ensemble optimization for time series cryptocurrency prediction

Cryptocurrency is an advanced digital currency that is secured by encryption, making it nearly impossible to forge or duplicate. Many cryptocurrencies are blockchain-based with decentralized networks. The prediction of cryptocurrency prices is a very difficult task because of the absence of an appropriate analytical basis to substantiate their claims. Cryptocurrencies are also dependent on several variables, such as technical advancement, internal competition, market pressure, economic concerns, security, and political considerations. This paper proposed the hybrid walk-forward ensemble optimization technique and applied it to predict the daily prices of fifteen cryptocurrencies, such as Cardano (ADA-USD), Bitcoin (BTC-USD), Dogecoin (DOGE-USD), Ethereum Classic (ETC-USD), Chainlink (LINK-USD), Litecoin (LTC-USD), NEO (NEO-USD), Tron (TRX-USD), Tether (USDT-USD), NEM (XEM-USD), Stellar (XLM-USD), Ripple (XRP-USD), and Tezos (XTZ-USD). A performance comparison of these cryptocurrencies was done using classical statistical models, machine learning algorithms, and deep learning algorithms on different cryptocurrency time series. Simulation results show that our proposed model performed better in terms of cryptocurrency prediction accuracy compared to the classical statistical model and machine and deep learning algorithms used in this paper.

Clustering and Modelling of the Top 30 Cryptocurrency Prices Using Dynamic Time Warping and Machine Learning Methods

Cryptocurrencies are a new field of investment opportunities that has experienced a significant growth in the last decade. The crypto market was capitalized at more than USD 3000 bn, having grown from USD 10 m over the period 2011–2021. Generating high returns, investments in cryptocurrencies have also shown high levels of price volatility. By comparing the performance of cryptocurrencies (measured by the crypto index) and standard equities (included in the S&P 500 index), we found that the former has outperformed the latter 14 times over the last two years. In the present paper, we analyzed the 2012–2022 global crypto market developments and main constituents. With a focus on the top 30 cryptocurrencies and their prices, as of 9 April 2022, covering data of the two major market stress events—outbreaks of the COVID-19 pandemic (February 2020) and the Russian invasion of Ukraine (February 2022). We applied the dynamic time warping method including barycentre averaging and k-Shape clustering of time series. The use of the dynamic time warping has been essential for the preparation of data for subsequent clustering and forecasting. In addition, we compared performance of cryptocurrencies and equities. Cryptocurrency time series are rather short, sometimes involving high levels of volatility and including multiple data gaps, whereas equity time series are much longer and well-established. Identifying similarities between them allows analysts to predict crypto prices by considering the evolution of similar equity instruments and their responses to historical events and stress periods. Moreover, we tested various forecasting methods on the 30 cryptocurrencies to compare traditional econometric methods with machine learning approaches.

Time-Transformed Test for Bubbles under Non-stationary Volatility

Abstract This paper is devoted to testing for bubbles under time-varying non-stationary volatility. Because the limiting distribution of the seminal Phillips, Wu, and Yu (2011) test depends on the variance function and usually requires a bootstrap implementation under heteroskedasticity, we construct the test based on a deformation of the time domain. The proposed test is asymptotically pivotal under the null hypothesis and its limiting distribution coincides with that of the standard test under homoskedasticity, so that the test does not require computationally extensive methods for inference. Appealing finite sample properties are demonstrated through Monte-Carlo simulations. An empirical application demonstrates that the upsurge behavior of cryptocurrency time series in the middle of the sample is partially explained by the volatility change.

Comparative Performance of Machine Learning Ensemble Algorithms for Forecasting Cryptocurrency Prices

This paper discusses the problems of short-term forecasting of cryptocurrency time series using a supervised machine learning (ML) approach. For this goal, we applied two of the most powerful ensemble methods including Random Forests (RF) and Stochastic Gradient Boosting Machine (SGBM). As the dataset was collected from daily close prices of three of the most capitalized coins: Bitcoin (BTC), Ethereum (ETH) and Ripple (XRP), and as features we used past price information and technical indicators (moving average). To check the effectiveness of these models we made an out-of-sample forecast for selected time series by using the one step ahead technique. The accuracy rate of the forecasted prices by using RF and GBM were calculated. The results verify the applicability of the ML ensembles approach for the forecasting of cryptocurrency prices. The out of sample accuracy of short-term prediction daily close prices obtained by the SGBM and RF in terms of Mean Absolut Percentage Error (MAPE) for the three most capitalized cryptocurrencies (BTC, ETH, and XRP) were within 0.92-2.61 %.

Ensemble Deep Learning Models for Forecasting Cryptocurrency Time-Series

Nowadays, cryptocurrency has infiltrated almost all financial transactions; thus, it is generally recognized as an alternative method for paying and exchanging currency. Cryptocurrency trade constitutes a constantly increasing financial market and a promising type of profitable investment; however, it is characterized by high volatility and strong fluctuations of prices over time. Therefore, the development of an intelligent forecasting model is considered essential for portfolio optimization and decision making. The main contribution of this research is the combination of three of the most widely employed ensemble learning strategies: ensemble-averaging, bagging and stacking with advanced deep learning models for forecasting major cryptocurrency hourly prices. The proposed ensemble models were evaluated utilizing state-of-the-art deep learning models as component learners, which were comprised by combinations of long short-term memory (LSTM), Bi-directional LSTM and convolutional layers. The ensemble models were evaluated on prediction of the cryptocurrency price on the following hour (regression) and also on the prediction if the price on the following hour will increase or decrease with respect to the current price (classification). Additionally, the reliability of each forecasting model and the efficiency of its predictions is evaluated by examining for autocorrelation of the errors. Our detailed experimental analysis indicates that ensemble learning and deep learning can be efficiently beneficial to each other, for developing strong, stable, and reliable forecasting models.

Financial Bubbles: A Study of Co-Explosivity in the Cryptocurrency Market

Cryptocurrencies have recently captured the interest of the econometric literature, with several works trying to address the existence of bubbles in the price dynamics of Bitcoins and other cryptoassets. Extremely rapid price accelerations, often referred to as explosive behaviors, followed by drastic drops pose high risks to investors. From a risk management perspective, testing the explosiveness of individual cryptocurrency time series is not the only crucial issue. Investigating co-explosivity in the cryptoassets, i.e., whether explosivity in one cryptocurrency leads to explosivity in other cryptocurrencies, allows indeed to take into account possible shock propagation channels and improve the prediction of market collapses. To this aim, our paper investigates the relationships between the explosive behaviors of cryptocurrencies through a unit root testing approach.

Topological recognition of critical transitions in time series of cryptocurrencies

We analyze four major cryptocurrencies (Bitcoin, Ethereum, Litecoin, and Ripple) before the digital asset market crash at the beginning of 2018. We also analyze Bitcoin before some of the mini-crashes that occurred during the period 2016–2018. All relevant time series exhibited a highly erratic behavior.We introduce a methodology that combines topological data analysis with a machine learning technique – k-means clustering – in order to characterize the emerging chaotic regime in a complex system approaching a critical transition.We first test our methodology on the complex system dynamics of a Lorenz-type attractor. Then we apply it to the four major cryptocurrencies. We find early warning signals for critical transitions, i.e., crashes, in the cryptocurrency markets.

Portfolio value-at-risk with two-sided Weibull distribution: Evidence from cryptocurrency markets

This paper extends the univariate two-sided Weibull distribution to a multivariate case for portfolio-value-at-risk estimation. This method allows to capture the stylized facts of the time series of cryptocurrencies, such as extreme volatility, volatility clustering, very heavy tails, and skewness. This new portfolio risk model is applied to a cryptocurrency portfolio consisting of four major coins: Bitcoin, Litecoin, Ripple, and Dash. The predictive performance of the proposed model is compared with several widely used models. We find that the portfolio value-at-risk with two-sided Weibull distribution outperforms the other models.

Comparing the Forecasting of Cryptocurrencies by Bayesian Time-Varying Volatility Models

This paper studies the forecasting ability of cryptocurrency time series. This study is about the four most capitalised cryptocurrencies: Bitcoin, Ethereum, Litecoin and Ripple. Different Bayesian models are compared, including models with constant and time-varying volatility, such as stochastic volatility and GARCH. Moreover, some cryptopredictors are included in the analysis, such as S&P 500 and Nikkei 225. In this paper, the results show that stochastic volatility is significantly outperforming the benchmark of VAR in both point and density forecasting. Using a different type of distribution, for the errors of the stochastic volatility, the student-t distribution is shown to outperform the standard normal approach.

Modelling multifractal properties of cryptocurrency market

The paper focuses on the study of the effect of long memory and the analysis of the multifractal properties of the time series of the most capitalized cryptocurrencies for the period from 2010 to 2018. To do this, the Hurst exponent is calculated by both R/S analysis and the Detrended Fluctuation Analysis being more stable in the case of non-stationary time series. Our results show that time series of cryptocurrencies to be persistent during almost the whole study period that do not allow accepting the hypothesis concerning the efficiency of the cryptocurrency market. We also found that (i) time series became anti-persistent during the periods of market crisis phenomena and turbulence; (ii) the Hurst exponents showed significant fluctuations about the value of 0.5. In addition, we conduct a multifractal analysis of cryptocurrency time series that allows us to assess the state and stability of the market.The calculated spectrum of multifractality shows that the cryptocurrency market comes out of a crisis state, since the width of the multifractality spectrum has the maximum value for all cryptocurrencies.

VOLATILITY SPILLOVERS AMONG THE CRYPTOCURRENCY TIME SERIES

This paper uses different multivariate GARCH models to model conditional correlations and analyze the volatility spillovers between cryptocurrency time series. The dynamic conditional correlation GARCH model is found to fit the data the best. Our empirical results are fourfold. First, on average, a $1 long position in BitShares (BTS) can be hedged for 15% with a short position in MonaCoin (MONA), while a $1 long position in MONA can be hedged for 14% with a short position in Ripple (XRP). Second, the average weight for the BTS/MONA portfolio is 0.48, indicating that for a $1 portfolio, 48% should be invested in BTS and 52% invested in MONA. Third, the average weight for the BTS/XRP portfolio indicates that 27% should be invested in BTS and 73 % invested in XRP. Finally, the average weight for the MONA/XRP portfolio indicates that 33% should be invested in MONA and 67% invested in XRP.

Trend Prediction Classification for High Frequency Bitcoin Time Series with Deep Learning

We provide a trend prediction classification framework named the random sampling method (RSM) for cryptocurrency time series that are non-stationary. This framework is based on deep learning (DL). We compare the performance of our approach to two classical baseline methods in the case of the prediction of unstable Bitcoin prices in the OkCoin market and show that the baseline approaches are easily biased by class imbalance, whereas our model mitigates this problem. We also show that the classification performance of our method expressed as the F-measure substantially exceeds the odds of a uniform random process with three outcomes, proving that extraction of deterministic patterns for trend classification, and hence market prediction, is possible to some degree. The profit rates based on RSM outperformed those based on LSTM, although they did not exceed those of the buy-and-hold strategy within the testing data period, and thus do not provide a basis for algorithmic trading.

Forecasting cryptocurrencies under model and parameter instability

This paper studies the predictability of cryptocurrency time series. We compare several alternative univariate and multivariate models for point and density forecasting of four of the most capitalized series: Bitcoin, Litecoin, Ripple and Ethereum. We apply a set of crypto-predictors and rely on dynamic model averaging to combine a large set of univariate dynamic linear models and several multivariate vector autoregressive models with different forms of time variation. We find statistically significant improvements in point forecasting when using combinations of univariate models, and in density forecasting when relying on the selection of multivariate models. Both schemes deliver sizable directional predictability.

On generalized bivariate student-t Gegenbauer long memory stochastic volatility models with leverage: Bayesian forecasting of cryptocurrencies with a focus on Bitcoin

A Gegenbauer long memory stochastic volatility model with leverage and a bivariate Student’s t-error distribution to model the innovations of the observation and latent volatility jointly for cryptocurrency time series is presented. This is inspired by the deep rooted characteristics found in cryptocurrencies. Until recently their econometric properties have not been thoroughly investigated. Thus, a rigorous in-sample simulation is conducted to assess the performance of the model with its nested alternatives and study the behavior of many cryptocurrencies and in particular Bitcoin. The data analysis is initiated with a broad scope of 114 cryptocurrencies, then a more detailed understanding of five of the most popular cryptocurrencies and followed up with forecasts focused specifically on Bitcoin (while other forecasts are available as supplementary material). The model parameters are estimated with Bayesian approach using Markov Chain Monte Carlo sampling. In order to implement model selection, the Deviance Information Criterion (DIC) is used. Proposed models are compared with many popular models including those commonly used in industry. The models are applied in a Value-at-Risk (VaR) context and several measures are used to assess model performance.