Abstract We investigate the out-of-sample diversification benefits of cryptocurrencies from a generalised perspective, a cryptocurrency-factor level, with traditional and machine-learning-enhanced asset allocation strategies. The cryptocurrency factor portfolios are formed in an analogous way to equity anomalies by using more than 2000 cryptocurrencies. The findings indicate that a stock–bond portfolio incorporating size- and momentum-based cryptocurrency factors can achieve statistically significant out-of-sample diversification benefits for investors with different risk preferences. Additionally, machine-learning-enhanced asset allocation strategies can boost the traditional approaches by enriching (shrinking) the distributions of weights allocated to potentially effective cryptocurrency factors. Our findings are robust to (i) the inclusion of transaction costs, (ii) an alternative benchmark portfolio, and (iii) a rolling-window estimation scheme.
Fabian E. Eska, Yanghua Shi, Erik Theissen, Marliese Uhrig‐Homburg
This paper examines the impact of cryptocurrency design features on their return volatility. We compile a sample of 58 cryptocurrencies, adopt the taxonomy of design features proposed by Eska et al. (2022), and estimate LASSO regressions. We document that older cryptocurrencies tend to be less volatile. Networks with mandatory transaction fees, cryptocurrencies based on (delegated) Proof-of-Stake, and those developed by private for-profit entities tend to be more volatile. Furthermore, we provide evidence that networks passing transaction fees and/or tips on to verifiers are associated with higher volatility levels.
Ethereum has recently surged in popularity, as it can hold various digital tokens and decentralized applications.This paper aims to predict UNI's price in USD through dynamic network analysis and time-series analysis.Previous research in this field rarely considers comprehensive network analysis while predicting token price.This paper puts forward a strengthened Bidirectional LSTM model that includes token economical features and network features.We use Root Mean Squared Error (RMSE) to verify the validity and compare it with other LSTM and GRU models on performance.Lastly, a logarithm difference method for data preprocessing was introduced to resolve the lag problems.
Muhammad Anas, Syed Jawad Hussain Shahzad, Larisa Yarovaya
Abstract As the crypto-asset ecosystem matures, the use of high-frequency data has become increasingly common in decentralized finance literature. Using bibliometric analysis, we characterize the existing cryptocurrency literature that employs high-frequency data. We highlighted the most influential authors, articles, and journals based on 189 articles from the Scopus database from 2015 to 2022. This approach enables us to identify emerging trends and research hotspots with the aid of co-citation and cartographic analyses. It shows knowledge expansion through authors’ collaboration in cryptocurrency research with co-authorship analysis. We identify four major streams of research: (i) return prediction and measurement of cryptocurrency volatility, (ii) (in)efficiency of cryptocurrencies, (iii) price dynamics and bubbles in cryptocurrencies, and (iv) the diversification, safe haven, and hedging properties of Bitcoin. We conclude that highly traded cryptocurrencies’ investment features and economic outcomes are analyzed predominantly on a tick-by-tick basis. This study also provides recommendations for future studies.
V. Akila, Nitin M.V.S., I S N V R Prasanth, S. R. M. A. Ayeshmi M. · 5 authors
Cryptocurrencies have gained immense popularity in recent years as an emerging asset class, and their prices are known to be highly volatile. Predicting cryptocurrency prices is a difficult task due to their complex nature and the absence of a central authority. In this paper, our proposal is to employ Long Short-Term Memory (LSTM) networks, a type of deep learning technique to forecast the prices of cryptocurrencies. We use historical price data and technical indicators as inputs to the LSTM model, which learns the underlying patterns and trends in the data. To improve the accuracy of the predictions, we also incorporate a Change Point Detection (CPD) technique using the Pruned Exact Linear Time (PELT) algorithm. This method allows us to detect significant changes in cryptocurrency prices and adjust the LSTM model accordingly, leading to better predictions. We evaluate our approach predominantly on Bitcoin cryptocurrency, but the model can be implemented on other cryptocurrencies provided there are valid historical price data. Our experimental results show that our proposed model outperforms the baseline LSTM algorithm, achieving higher accuracy and better performance in terms of Mean Absolute Error (MAE), Mean Square Error (MSE), and Root Mean Square Error (RMSE). Our research findings suggest that combining deep learning techniques such as LSTM with change point detection techniques such as PELT can improve cryptocurrency price prediction accuracy and have practical implications for investors, traders, and financial analysts.
Christian Fieberg, Gerrit Liedtke, Thorsten Poddig, Thomas Walker · 5 authors
Abstract We propose CTREND, a new trend factor for cryptocurrency returns, which aggregates price and volume information across different time horizons. Using data on more than 3,000 coins, we employ machine learning methods to exploit information from various technical indicators. The resulting signal reliably predicts cryptocurrency returns. The effect cannot be subsumed by known factors and remains robust across different subperiods, market states, and alternative research designs. Moreover, it survives the impact of transaction costs and persists in big and liquid coins. Finally, an asset pricing model that incorporates CTREND outperforms competing factor models, providing a superior explanation of cryptocurrency returns.
We examine the volume-volatility relationship across Bitcoin futures and spot markets, using daily realised volatility measures estimated from high frequency intraday data. We estimate realised spot volatility across five major exchanges using both the standard volume weighted price and using a new approach, inspired by the CME Bitcoin Reference Rate methodology. We find that unexpected trading volume is the most important explanatory variable for BRR spot volatility, explaining 20% of variation in price volatility at exchange level. Conversely, we find that both expected and unexpected CME Bitcoin futures volumes play a very limited or even calming role in systemic volatility. Our findings suggest that CME Bitcoin futures are not independently contributing to systemic risk in Bitcoin over the period studied.
Deep learning methods have achieved significant success in various applications, including trend signal prediction in financial markets. However, most existing approaches only utilize price action data. In this paper, we propose a novel system that incorporates multiple data sources and market correlations to predict the trend signal of Ethereum cryptocurrency. We conduct experiments to investigate the relationship between price action, candlestick patterns, and Ethereum-Bitcoin correlation, aiming to achieve highly accurate trend signal predictions. We evaluate and compare two different training strategies for Convolutional Neural Networks (CNNs), one based on transfer learning and the other on training from scratch. Our proposed 1-Dimensional CNN (1DCNN) model can also identify inflection points in price trends during specific periods through the analysis of statistical indicators. We demonstrate that our model produces more reliable predictions when utilizing multiple data representations. Our experiments show that by combining different types of data, it is possible to accurately identify both inflection points and trend signals with an accuracy of 98%.
Ahmet Faruk Aysan, Massimiliano Caporin, Oğuzhan Çepni
This paper analyzes the relationship between price jumps and news sentiment in cryptocurrencies. We detect jumps at the intraday level and correlate their occurrence with sentiment-related events through logistic regressions. We show that the release of information increases the probability of price jumps. By examining the content of news stories, we find that sentiment dimensions limited to emotions or related to market fundamentals have more potential to result in price jumps than others, suggesting that “words are not all created equal”. Jump sensitivity to news sentiment varies across different coin characteristics.
The Ethereum blockchain network enables transaction processing and smart-contract execution through levies of transaction fees, commonly known as gas fees. This framework mediates economic participation via a market-based mechanism for gas fees, permitting users to offer higher gas fees to expedite processing. Historically, the ensuing gas fee volatility led to critical disequilibria between supply and demand for block space, presenting stakeholder challenges. This study examines the dynamic causal interplay between transaction fees and economic subsystems leveraging the network. By utilizing data related to unique active wallets and transaction volume of each subsystem and applying time-varying Granger causality analysis, we reveal temporal heterogeneity in causal relationships between economic activity and transaction fees across all subsystems. This includes (a) a bidirectional causal feedback loop between cross-blockchain bridge user activity and transaction fees, which diminishes over time, potentially signaling user migration; (b) a bidirectional relationship between centralized cryptocurrency exchange deposit and withdrawal transaction volume and fees, indicative of increased competition for block space; (c) decentralized exchange volumes causally influence fees, while fees causally influence user activity, although this relationship is weakening, potentially due to the diminished significance of decentralized finance; (d) intermittent causal relationships with maximal extractable value bots; (e) fees causally influence non-fungible token transaction volumes; and (f) a highly significant and growing causal influence of transaction fees on stablecoin activity and transaction volumes highlight its prominence. These results inform strategic considerations for stakeholders to more effectively plan, utilize, and advocate for economic activities on Ethereum, enhancing the understanding and optimization of within the rapidly evolving economy.