Jacques Fleischer, Gregor von Laszewski, Carlos Theran, Yohn Jairo Parra Bautista
In this paper we apply neural networks and Artificial Intelligence (AI) to historical records of high-risk cryptocurrency coins to train a prediction model that guesses their price. This paper's code contains Jupyter notebooks, one of which outputs a timeseries graph of any cryptocurrency price once a CSV file of the historical data is inputted into the program. Another Jupyter notebook trains an LSTM, or a long short-term memory model, to predict a cryptocurrency's closing price. The LSTM is fed the close price, which is the price that the currency has at the end of the day, so it can learn from those values. The notebook creates two sets: a training set and a test set to assess the accuracy of the results. The data is then normalized using manual min-max scaling so that the model does not experience any bias; this also enhances the performance of the model. Then, the model is trained using three layers -- an LSTM, dropout, and dense layer-minimizing the loss through 50 epochs of training; from this training, a recurrent neural network (RNN) is produced and fitted to the training set. Additionally, a graph of the loss over each epoch is produced, with the loss minimizing over time. Finally, the notebook plots a line graph of the actual currency price in red and the predicted price in blue. The process is then repeated for several more cryptocurrencies to compare prediction models. The parameters for the LSTM, such as number of epochs and batch size, are tweaked to try and minimize the root mean square error.
Purpose Perhaps the most popular pricing model among Bitcoin enthusiasts is the stock-to-flow (S2F) model. The model gained significant traction after successfully predicting the meteoric rise of Bitcoin prices from late 2020 to early 2021. This paper dissects the S2F model for Bitcoin empirically to determine its viability and investigate whether investors can profit from an S2F-based trading strategy. Design/methodology/approach This paper, dissects the S2F model for Bitcoin by putting it through a battery of tests to examine its design, characteristics, robustness and appropriateness. Findings Overall, this paper finds the S2F model to be insensitive to differing assumptions in the early stages of the model, alleviating concerns about data mining. This paper produces a dynamic S2F model with no peek-ahead bias and shows evidence that prediction accuracy increases over time. Finally, this paper shows that a dynamic trading strategy that goes long (short) when Bitcoin is undervalued (overvalued) according to S2F is far less profitable than a classic buy-and-hold strategy. Originality/value To the best of the authors’ knowledge, this is the first paper to analyze the S2F model in an academic setting by providing a rigorous assessment of the model's construction. This paper demonstrates how the model can be implemented realistically without the peek-ahead bias, creating a tool that can be used contemporaneously by investors.
The purpose of our study is to figure out the transitions of the cryptocurrency market due to the outbreak of COVID-19 through network analysis, and we studied the complexity of the market from different perspectives. To construct a cryptocurrency network, we first apply a mutual information method to the daily log return values of 102 digital currencies from January 1, 2019, to December 31, 2020, and also apply a correlation coefficient method for comparison. Based on these two methods, we construct networks by applying the minimum spanning tree and the planar maximally filtered graph. Furthermore, we study the statistical and topological properties of these networks. Numerical results demonstrate that the degree distribution follows the power-law and the graphs after the COVID-19 outbreak have noticeable differences in network measurements compared to before. Moreover, the results of graphs constructed by each method are different in topological and statistical properties and the network's behavior. In particular, during the post-COVID-19 period, it can be seen that Ethereum and Qtum are the most influential cryptocurrencies in both methods. Our results provide insight and expectations for investors in terms of sharing information about cryptocurrencies amid the uncertainty posed by the COVID-19 pandemic.
In recent years, cryptocurrencies have become a new topic for financial studies. In this study, the effects of positive and negative events related to cryptocurrencies on the prices of related cryptocurrencies were researched using the event study. These events include major listing, delisting and airdrop announcements, and SEC enforcements. As a result of the analysis, 22 significant abnormal return values related to negative events and eight significant abnormal return values related to positive events were determined at 1% significance level within the event window (-5, +10). Therefore, it has been determined that negative events have more effect on cryptocurrencies than positive events. The number of significant cumulative abnormal return values obtained (13 for negative events, three for positive events) also supports these results. The results of the study have crucial implications for investors, centralized cryptocurrency exchanges, and cryptocurrency CEOs. Even after the negative events were announced publicly, pull out of the market will prevent investors from making more losses. In addition, it is recommended that investors sell for profits in case of a rapid high return on the day of the listing announcement. Because it was determined that the prices returned to the equilibrium prices at the closing.
In this paper, we will consider a governance token pricing algorithm that conducts liquidity operations on AMM (CPMM) DEXs (automated market maker decentralized exchanges) with liquidity that belongs to a decentralized autonomous organization (DAO), also called protocol-owned liquidity (POL). The primary aim of the protocol is maintaining a price peg by determining algorithmically when and how to carry out interventions that consist of two steps: extracting liquidity from an AMM liquidity pool and conducting "token swap" operations. We will cover setting up an optimal peg function as a weighted sum of certain normalized factors, which are to be determined collectively by the DAO. In particular, we will review various arithmetic invariants of liquidity intervention, which brings the price to a peg while leaving total liquidity intact, and show how such interventions can be substituted in practice by so-called PMM (proactive market maker) protocols.
Zeyd Boukhers, Azeddine Bouabdallah, Cong Yang, Jan Jürjens
Since Bitcoin first appeared on the scene in 2009, cryptocurrencies have become a worldwide phenomenon as important decentralized financial assets. Their decentralized nature, however, leads to notable volatility against traditional fiat currencies, making the task of accurately forecasting the crypto-fiat exchange rate complex. In this study, we examine the various independent factors that affect the Bitcoin-Dollar exchange rate's volatility. To this end, we propose CoMForE, a multimodal AdaBoost-LSTM ensemble model, which not only utilizes historical trading data but also incorporates public sentiments from related tweets, public interest demonstrated by search volumes, and blockchain hash-rate data. Our developed model goes a step further by predicting fluctuations in the overall cryptocurrency value distribution, thus increasing its value for investment decision-making. We have subjected this method to extensive testing via comprehensive experiments, thereby validating the importance of multimodal combination over exclusive reliance on trading data. Further experiments show that our method significantly surpasses existing forecasting tools and methodologies, demonstrating a 19.29% improvement. This result underscores the influence of external independent factors on cryptocurrency volatility.
This study attempts to answer the question: “Is the Cryptocurrency Policy Uncertainty (UCRY Policy) a determinant of the Bitcoin’s price (BTC)?”. Besides, this study uses these factors as explanatory variables for the BTC movements alongside the UCRY Policy and control variables such as the velocity of the Bitcoin in circulation (BC), the computational power of Bitcoin (HR), popularity (PO), and exchange rate (EX). In the study, December 30, 2013- February 21, 2021, was determined as the term and weekly data were investigated. The ARDL bounds testing method was used to determine the relationship between the variables. According to empirical findings, this study suggests that the UCRY Policy is essential to the BTC. There is a negative relationship between UCRY Policy and BTC. When UCRY Policy increases, BTC decreases, holding other variables constant. Besides, this study shows that control variables can be used as determinants of BTC. In long run, BC and HR have a significant, positive relationship with BTC. The EX has a significant, negative relationship with BTC. The PO has a significant, positive relationship with BTC in the short run. In addition, this study demonstrates that UCRY Policy can be used as a type of uncertainty index for Bitcoin.
This paper proposes a novel asymmetric jump model for modeling interactions in discontinuous movements in asset prices. Given the jump behavior and high volatility levels in cryptocurrency markets, we apply our model to cryptocurrencies to study the impact of various types of jumps occurring in one cryptocurrency’s price process on the discontinuity component of the realized volatility of other cryptocurrencies. Our model also allows us to assess the impact of co-jumps. Using high-frequency data to compute the daily realized volatility, we show that downside, upside, and small jumps observed in cryptocurrencies negatively affect the jump component of other cryptocurrencies’ realized volatility, while large jumps have the opposite effect. We further find significant asymmetric effects between small and large as well as between downside and upside jumps for several cryptocurrencies. Moreover, we find evidence of co-jumping behavior, which can trigger future jumps. The practical implications of our findings are also discussed. Finally, we extend our analysis to study the effects of jumps in mainstream financial assets on cryptocurrencies’ jump behavior and find that upside and downside jumps observed in the S&P 500 index negatively impact cryptocurrency jumps.
Abstract The annual energy consumption of cryptocurrencies has been increasing in recent years. This paper studies the cryptocurrencies return volatility spillover and the underlying dynamics of five cryptocurrencies, namely Bitcoin, Bitcoin Cash, Ethereum, Ripple XRP and Litecoin's impact on four energy markets, namely Nifty Energy Index, S&P 500 Energy Index, S&P/TSX Canadian Energy Index and Shanghai Stock Exchange Energy Index for the period 2016–2021. We employed the Granger Causality Test and DCC MGARCH model to investigate the integration between cryptocurrencies and the energy markets. From the empirical analyses, we find that the overall time‐varying correlation between cryptocurrencies and the energy markets is low and weak. This study may be helpful for investors, academia and policymakers.
Werner Kristjanpoller, Leonardo H.S. Fernandes, Benjamin Miranda Tabak
Cryptocurrencies play a pivotal role in the financial market. Given this, we perform the asymmetric multifractal cross-correlation analysis to examine the weak form of the Efficient Market Hypotheses (EMH) considering two temporal scales. In the daily scale, we find that the pair Bitcoin–Litecoin displays the largest multifractal spectrum. While, in the hourly scale, the pair Bitcoin–Ethereum presents the largest multifractal spectrum. Our empirical evidence has rejected the weak form of the EMH and clearly suggests that the dynamics of the analyzed cryptocurrency pairs are in line with the Fractal Market Hypothesis (FMH). Cross-correlation asymmetries are more persistent for small fluctuations than for large fluctuations. The results are essential for investors, portfolio and risk managers, and policymakers.
Kaos Teorisi, doğrusal olmayan dinamik sistemlerin davranışlarını tanımlar ve ekonomi alanında pek çok verinin modellenmesinde kullanılır. Kaos teori, sistemin doğrusal olmayan ve deterministik bir süreç olduğu varsayımlarına dayanır. Doğrusal modeller, ekonometrik sistemleri karmaşıklıklarını ortaya çıkarmakta yetersiz kalmaktadır. Bu çalışmanın amacı, Bitcoin günlük fiyatlarının zamana bağlı doğrusal olmayan dinamik bir sistem tarafından üretilip üretilmediğini araştırmak ve sistemin uzun vadede geleceğe yönelik tahmin yeteneğini araştırmak ve bir tahminleme modeli oluşturmaktır. Birçok ekonomik veri serisinin kaotik davranış gösterdiği bilinmektedir. Bu çalışmada, Bitcoin fiyatlarının kaotik yapısı incelenmiş ve regresyon yöntemi kullanılarak tahmin modeli kurulmuştur. Diğer bir ifadeyle amaç, Bitcoin fiyatlarının getirilerinin kaotik bir davranış gösterip göstermediğini ortaya koyarak elde edilen gömme (embedding) boyutuna bağlı olarak regresyon yöntemini kullanarak tahmin modeli oluşturmaktır. Çalışmada, 2021 Şubat – 2021 Kasım döneminde günlük kapanış fiyatı ( $ ) veri olarak kullanılmıştır. (URL-1,2021)
This paper presents how volatility propagates through the cryptocurrency market. Our paper provides evidence for volatility connectedness on cryptocurrencies. The different econometric techniques, including the stochastic volatility (SVOL) model and time-varying parameter VAR models using a quasi-Bayesian local likelihood (QBLL), are applied to measure the volatility of the cryptocurrency market. Using high-frequency, intra-day data of the largest cryptocurrencies over 2018–2021, we detect the great volatility of the cryptocurrency market are the beginning of 2019, the beginning of 2020, and throughout the year of 2021. The total connectedness values suggest that the cryptocurrency market becomes volatile as the new strains of the COVID-19 appear at the end of 2021. However, by using directional connectedness, we reveal that there are negative and positive spillovers from a specific cryptocurrency to other cryptocurrencies. The great fluctuations in the period before the COVID-19 health crisis stem from the positive resonance (symmetric) between the volatility of each cryptocurrency, while this health crisis leads to substantially positive and negative spillovers (asymmetric) of cryptocurrencies, and this makes market volatility weaker than it actually is.
Investors now have a multitude of non-traditional assets to choose from, especially from the spectrum of alternative assets, such as financial digital assets. We start from the premise that there is a high risk associated with investing in financial digital assets, along with the opportunities presented from these emerging digital markets that evolve in a decentralized environment. We will be looking at the two major digital assets, specifically Bitcoin (BTC) and Ethereum (ETH), as per their dominance within the markets of crypto assets. This paper will focus on the evolution of financial digital assets and the impact on portfolio assessment that have allocations for BTC and ETH. In order to identify the value and potential of these financial digital assets, we will be addressing volatility and portfolio risks by means of a Vector Autoregression model on the returns of both, BTC and ETH.
We use a time-varying vector autoregressive model to investigate the dynamic effect of investor attention on Bitcoin speculation and then examine the association of this effect with five types of events in the Bitcoin market. The results indicate that investor attention has a positive effect on Bitcoin speculation and this effect changes with time and decays as lag phases increase. Policy-related events are the key factors that make this effect time-varying, while safety events have no obvious impact. Besides, our results find the existence of contrarian strategy in the Bitcoin market.
We leverage a transaction costs narrative to provide a theoretically unified presentation of the evolution of exchange, with the latest evolutionary frontier being cryptocurrency and decentralized finance. We show that with each new development in the evolution of money, the new form or medium of exchange must reduce transaction costs relative to relevant alternatives. The development of blockchain and cryptocurrency reduced the cost of transferring currency by removing the need for a trusted third party to intermediate funds while also providing the benefit of anonymity/pseudonymity. Likewise, decentralized finance does not require a third party to intermediate savings and investment and can provide contingent anonymity to borrowers. While these innovations have attracted investment in the economically developed world, they appear to have significantly reduced transaction costs for transactors who might otherwise be defrauded of funds by corrupt governments who may extort third parties responsible for intermediating funds.