Matthias Schnaubelt, Jonas Rende, Christopher Krauß
The majority of electronic markets worldwide employ limit order books, and the recently emerging exchanges for cryptocurrencies pose no exception. With this work, we empirically analyze whether commonly observed empirical properties from established limit order exchanges transfer to the cryptocurrency domain. Based on the literature, we establish a structured methodological framework to conduct analyses in a systematic and comprehensive way. We then present results from a unique and extensive limit order data set acquired from major cryptocurrency exchanges for the currency pair Bitcoin to US Dollar. We recover many observations from mature markets, such as a symmetry between the average ask and the average bid side of the order book, autocorrelation in returns on the smallest time scales only, volatility clustering and the timing of large trades. We also observe some idiosyncrasies: The distributions of trade size and limit order prices deviate from commonly observed patterns. Also, we find limit order books to be relatively shallow and liquidity costs to be relatively high when compared to established markets.
Matheus José Silva de Souza, Fahad Almudhaf, Bruno Miranda Henrique, Ana Beatriz Silveira Negredo · 7 authors
This paper aims to investigate how Machine Learning (ML) techniques perform in the prediction of cryptocurrency prices. We answer if Support Vector Machines (SVM) and Artificial Neural Networks (ANN) based strategies can generate abnormal risk-adjusted returns when applied to Bitcoin, the largest decentralized digital currency in terms of market capitalization. Findings indicate that traders are able to earn conservative returns on the risk adjusted basis, even accounting for transaction costs, when using SVM. Furthermore, the study suggests that ANN can explore short run informational inefficiencies to generate abnormal profits, being able to beat even buy-and-hold during strong bull trends.
Abstract Using extreme value analysis, we investigate the tail risk behavior of the high‐frequency (hourly) log returns of four most popular cryptocurrencies. The analysis is conducted on high‐frequency returns data, estimating value at risk and expected shortfall with varying thresholds. We find that Ripple is the most risky cryptocurrency exhibiting the largest potential gain or loss for both positive and negative (hourly) log returns at every percentile and threshold. Bitcoin is the least risky cryptocurrency.
Recent research on the economics of digitization investigates the dramatic changes in markets by digital technology. Digital technology has caused significant differences in in the cost of storage, computation, and transmission of data. As one of the latest sign of digitization in our life, the use of cryptocurrencies has been drawing attention of all market players. Blockchain technology is recently reallocating resources, restructuring of routines, changing market relationships and patterns of the flow of goods and services. This study investigates the coherence of Bitcoin with the movements of the main indicators of different markets. Aim of this research is to clarify whether this cryptocurrency follows the market conditions or not. Because the movements of the market values of Bitcoin are aimed to be investigated to show that they can be used separately as a tool of risk management. For this reason, wavelet analysis has been employed to define cross-correlation between time series of the daily USD value of Bitcoin and some market indicators. Some literature asserts that Bitcoin has recently started to follow market conditions. If Bitcoin process follows the market conditions more than before, that if This analysis will explain if there is a change in hedging possibility of Bitcoin recently.
In this study, we investigate the relationship between Bitcoin mining technology variables and Bitcoin returns, using a GARCH-M model. Additionally, we examine the predictive power of the mining technology variables on future Bitcoin returns. We find that mining difficulty and block size are inversely related to Bitcoin returns. Additionally, our findings signifying that the higher the block size the lower the Bitcoin price and consequently the lower the expected return. Second, our findings show that mining difficulty and block size are robust predictors of future Bitcoin returns.
The Google Trends 1 search analysis service and the Telegram 2 messaging platform are investigated to determine their respective relationships to cryptocurrency price behaviour. It is shown that, in contrast to earlier findings, the relationship between cryptocurrency price movements and internet search volumes obtained from Google Trends is no longer consistently positive, with strong negative correlations detected for Bitcoin and Ethereum during June 2018. Sentiment extracted from cryptocurrency investment groups on Telegram is found to be positively correlated to Bitcoin and Ethereum price movements, particularly during periods of elevated volatility. The number of messages posted on a Bitcoin-themed Telegram group is found to be an indicator of Bitcoin price action in the subsequent week. A long shortterm memory (LSTM) recurrent neural network is developed to predict the direction of cryptocurrency prices using data obtained from Google Trends and Telegram. It is shown that Telegram data is a better predictor of the direction of the Bitcoin market than Google Trends. The converse is true for Ethereum. The LSTM model produces the most accurate results when predicting price movements over a one-week period.
Daniel Traian Pele, Miruna Mazurencu-Marinescu-Pele
In this paper we investigate the ability of several econometrical models to forecast value at risk for a sample of daily time series of cryptocurrency returns. Using high frequency data for Bitcoin, we estimate the entropy of intraday distribution of logreturns through the symbolic time series analysis (STSA), producing low-resolution data from high-resolution data. Our results show that entropy has a strong explanatory power for the quantiles of the distribution of the daily returns. Based on Christoffersen's tests for Value at Risk (VaR) backtesting, we can conclude that the VaR forecast build upon the entropy of intraday returns is the best, compared to the forecasts provided by the classical GARCH models.
OlaOluwa S. Yaya, Ahamuefula E. Ogbonna, Robert Mudida, Nuruddeen Abu
Abstract This article investigates both market efficiency and volatility persistence in 12 cryptocurrencies during pre‐crash and post‐crash periods. The article contributes to the debate on the market efficiency of cryptocurrencies in the presence of volatility, considering robust fractional integration methods in both linear and nonlinear setups. We find that markets of Bitcoin and most altcoins considered in our study can be dubbed as efficient, and are also highly volatile, particularly, in the post‐crash period that we are experiencing now. The volatilities are more likely to persist for a shorter period than volatilities in the pre‐crash period. Our work, therefore, renders important information to cryptocurrency market participants and portfolio managers.
Muhammad Ali Nasir, Toan Luu Duc Huynh, Sang Phu Nguyen, Duy Duong
In the context of the debate on the role of cryptocurrencies in the economy as well as their dynamics and forecasting, this brief study analyzes the predictability of Bitcoin volume and returns using Google search values. We employed a rich set of established empirical approaches, including a VAR framework, a copulas approach, and non-parametric drawings, to capture a dependence structure. Using a weekly dataset from 2013 to 2017, our key results suggest that the frequency of Google searches leads to positive returns and a surge in Bitcoin trading volume. Shocks to search values have a positive effect, which persisted for at least a week. Our findings contribute to the debate on cryptocurrencies/Bitcoins and have profound implications in terms of understanding their dynamics, which are of special interest to investors and economic policymakers.
Purpose This study aims to compare investors of major conventional currencies and Bitcoin (BTC) investors by using the value at risk (VaR) method common risk measure. Design/methodology/approach The paper used a risk analysis named as VaR. The analysis has various computations that Historical Simulation and Monte Carlo Simulation methods were used for this paper. Findings Findings of the analysis are assessed in two different aspects of singular currency risk and portfolios built. First, BTC is found to be significantly risky with respect to the major currencies; and it is six times riskier than the singular most risky currency. Second, in terms of inclusion of BTC into a portfolio, which equally weights all currencies, it elevates overall portfolio risk by 98 per cent. Practical implications In spite of the remarkable risk level, it could be considered that investors are desirous of making an investment on BTC could mitigate their overall exposed risk relatively by building a portfolio. Originality/value The paper questions the risk level of Bitcoin, which is a digital currency. BTC, a matter of debate in the contemporary period, is seen as a digital currency free from control or supervision of a regulatory board. With the comparison of major currencies and BTC shows that how could be risky of a financial instrument without regulations. However, there is some advice for investors who would like to invest digital currencies despite the risk level in this study.
Bitcoin is a tremendously debated phenomenon in the world of finance and in recent the scientific literature on the topic has expanded. In this thesis,the bitcoin to US dollar exchange rate is examined through various conditional variance models to describe its highly volatile nature. We examine whether the introduction of bitcoin futurescontractsin late 2017 has had a decreasing impact on price volatility by estimatingthe unconditional variance. The log-return of the bitcoin exchange rate is analysed,and there is evidence of volatility clustering and time-varying volatility. Consequently, the variance is modelled through the GARCH(1,1), EGARCH(1,1) and GJR-GARCH(1,1) modelswith innovations followingthree distributions. The in-sample selection method selectedthe EGARCH(1,1) model where innovation terms follow a generalizederror distribution as the most parsimonious model. The findings show that volatilityhas not decreased after the introduction of bitcoin futures on regulated exchanges. \nKeywords: Bitcoin, conditional variancemodelling, bitcoin futures, price volatility exchange rate, statistical analysis
The aim of this paper is to examine the volatility spillover between bitcoin, gold and crude oil returns. (VAR) Model and three Multivariate GARCH Models (CCC-GARCH, BEKK-GARCH and DCC-GARCH) estimation techniques are applied using daily data from 1st January 2011 to August 31th, 2018. Further, these estimation results are used to analyze the relationship and the volatility spillovers among bitcoin and these commodity currencies. The findings reveal that the bidirectional spillover is confirmed between gold return and crude oil return. Low unidirectional spillover; from bitcoin return to gold return and from bitcoin to crude oil. We also notice that the DCC-GARCH model provides a better fit than the CCC-GARCH model and the BEKK-GARCH model. These findings have significant implications for both cryptocurrency these commodity currencies allocations and portfolio management.  JEL Classification Numbers: G10; G11; G58       Â
Zuzana Rakovská – Masaryk University, Faculty of Economics and Administration, Lipova 41a, 602 00 Brno, Czech Republic DOI: https://doi.org/10.31410/ITEMA.2019.241 3rd International Scientific Conference on Recent Advances in Information Technology, Tourism, Economics, Management and Agriculture – ITEMA 2019 – Bratislava, Slovakia, October 24, 2019, CONFERENCE PROCEEDINGS published by the Association of Economists and Managers […]
Bitcoin has emerged to become the most popular cryptocurrency and its presence\nhas the potential to disrupt existing payment and monetary systems. Over the past\ndecade, the bitcoin price has exhibited extreme volatility, puzzling for both academics\nand market practitioners. We examine the dynamic relationship between\ninvestor attention and the bitcoin price using principal component analysis and vector\nerror correction models and discover that investor attention is an important contributor\nin bitcoin price formation. Variance decomposition analysis suggests that\ninvestor attention explain a significant amount of future variations in the bitcoin\nprice, and investor attention can be used to predict direction of future price change.\nOur study offers insight into the bitcoin market and the economic impact of investor\nattention.
The purpose of this thesis is to learn if it’s possible to gain higher risk-adjusted profits than buy-and-hold -strategy on intraday bitcoin markets using moving averages or trading range breakout. Data used is price notations of bitcoin with 1-minute interval from 2017 to 2019. \n \nThere were many strategies that statistically significantly outperformed buy-and-hold -strategy even when trading fees were reduced. Different methods such as stop-loss and bands were able to significantly reduce volatility of returns. However, same trading rules don’t work well in different market conditions. Results from the train-set and test-set differed largely and were therefore not valid. Also, none of the strategies were able to outperform CCi30 cryptocurrency index when fees were reduced. \n \nTherefore, while some support was found to conclude that outperforming the index is possible, more data and research would be needed to validate these results.
The cryptocurrency literature has attempted to identifying factors that explain excess returns. We utilise principal component analysis to determine whether a (small) set of factors can explain returns and whether this varies over time. We find that a substantial proportion of cryptocurrency return variation is explained by a single principal component that is highly correlated with bitcoin returns. The explanatory power of this factor is greatest for larger cryptocurrencies and increases markedly in the most recent part of the sample. Our results have implications for investors determining optimal portfolio decisions and for policymakers wary of systemic risk.