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.
This paper analyses the efficiency of cryptocurrency markets by applying econometric models to different short-term investment horizons. A number of experiments are carried out to demonstrate that small training sets can still be used to build efficient and useful forecasts, which in turn can be transformed into straight-forward investment strategies. It also compares the application of selected models on cryptocurrency and mature stock markets. The forecasting accuracy of the models is explored using different error metrics and different horizons. The results suggest that the variation of the error estimates doesn’t appear to be tightly related to the maturity of the markets, but rather depends on the intrinsic characteristics of the analyzed time series.
In this study, a methodology is presented where a hybrid system combining an evolutionary algorithm with artificial neural networks (ANNs) is designed to make weekly directional change forecasts on the USD by inferring a prediction using closing spot rates of three currency pairs: EUR/USD, GBP/USD and CHF/USD. The forecasts made by the genetically trained ANN are compared to those made by a new variation of the simple moving average (MA) trading strategy, tailored to the methodology, as well as a random model. The same process is then repeated for the three major cryptocurrencies namely: BTC/USD, ETH/USD and XRP/USD. The overall prediction accuracy, uptrend and downtrend prediction accuracy is analyzed for all three methods within the fiat currency as well as the cryptocurrency contexts. The best models are then evaluated in terms of their ability to convert predictive accuracy to a profitable investment given an initial investment. The best model was found to be the hybrid model on the basis of overall prediction accuracy and accrued returns.
After the seminal work of Liu et al. (2015) finds that the realized volatility (RV) using 5-minute intervals performs well, economists tend to use this simple measure in applications. Existing literature in the cryptocurrency already relies on 5-minute RV, but no paper has evaluated whether 5-minute RV performs well compared with other realized measures. Following Liu et al. (2015), we show that the 5-minute RV of Bitcoin performs well compared to other realized measures, meaning that this result justifies the existing literature that already uses this simple measure. This paper also indicates that realized measures with longer intervals such as 120-minute RV could provide inaccurate estimates.
Cryptocurrencies are virtual currencies employed in blockchain transactions. They are particularly worthy of theoretical examination, given the limited academic literature on the subject. This paper constructs valuation models of bitcoin and altcoins, both as single investments and components of mutliple-asset portfolios. As single investments, cryptocurrencies are valued at the confluence of Legendre utility functions, with Esscher transformed Geometric Levy pricing processes. As part of portfolios, cryptocurrencies are contained in traditional Markowitz portfolios which are varied by increasing the proportion of the riskless asset, shorting the risky asset, or adding currency options. Theoretical formulations show that Markowitz models combined with bitcoin, located on the Capital Market Line (which we term CML portfolios), have low returns, mainly due to the presence of the riskless asset. Such portfolios are appropriately suited to the investment goals of risk-averse traders, while overlooking the preferences of risk-takers. To satisfy less risk-averse investors, we propose a high-return portfolio with 9 asset choices, consisting of risky assets, cryptocurrencies, US dollars, soybean futures, Treasury bond futures, oil futures, currency options on the US dollar, currency options on the Mexican peso, and technology, or biotechnology stocks. Laplace transforms are employed to suppress volatility, skewness, or kurtosis of returns, which empirical studies have found to contribute to tail risk contained in outliers in fat-tailed distributions.
One of the important tasks of every multi-asset portfolio managers is to assess how different asset classes interact with each other. Historical findings indicate that tradition risk asset classes exhibited various degrees of correlation, be they positive or negative, among each other. With the raise of crypto assets, such as bitcoin, it appears that crypto assets have gradually been considering as new investment class, at least from institutional aspect. This study reveals that the correlation of the digital currency with the longest price history, bitcoin, with other traditional assets is close to zero. Thus, from diversification point of view, this makes cryptocurrencies or bitcoin a perfectly uncorrelated asset which would benefit almost any portfolio. Further study is performed to investigate the cointegration relations among bitcoin and other asset classes. It is found that the spreads between bitcoin and some major tradition risk asset classes exist a mean reversion phenomenon. This enables asset managers to develop quantitative approaches for active management strategies. Models of cointegrated time series are common place in the literature and application in financial series. Correlation and cointegration are time series modelling techniques that have applied to financial markets. They are related but with different concepts. Correlation indicates co-movements in returns which is a short run measure requires frequent rebalancing to minimize losses, while cointegration measures long run tandem movements in prices to ensure long term performance for achieving returns. Two pairs of asset prices are found to have a common stochastic trend with stationary cointegrating vector, they are in theory considered for cointegration. This stochastic process displays a mean reversion in long run. If there exists a divergence in spread due to temporary shocks, one can expects to profit from performing pairs trading strategy by creating a short position on the outperforming one, at the same time with a long position on the underperforming one. In this study, trading signal would be generated for our pairs trading with bitcoin. Largely, our results empirically support over various asset classes during the period of estimation.
This present paper investigates day-of-the-week effect in some notable cryptocurrency in terms of pricing and market capitalizations. We applied fractional integration regression approach with dummies. We found non-significance of day-of-the-week effect in returns, while there is possible evidence of Monday and Friday effects in volatility of Bitcoin only. Non-significance of day-of-the-week effect in returns of Bitcoin and some other cryptocurrencies further support market efficiency of these markets.
Despite calls for regulation in the crypto utility token market, it is unclear how crypto token investors value current regulatory proposals. We find that on average, investors react negatively to news that increases the likelihood of securities and transparency-related regulation. We also find that this negative reaction is attenuated for tokens rated higher on quality and transparency by intermediaries, those that have higher levels of disclosure, and listed on more liquid exchanges. The observed variation in token transparency and this muted reaction suggest investors perceive disclosure costs to be lower for tokens in more transparent environments, suggesting that transparency matters to investors.
Yu-Min Lian, Chi-Hung Cheng, Shih-Hsun Lin, Jui-Hsuan Lin
In this study, we make use of both the specific method of Monte Carlo simulation and the spot-futures parity with the cost of carry to establish a dynamic price model of Bitcoin futures and to conduct the appraisals and numerical analyses. More specifically, the electricity fees and equipment costs are taken into account and the proposed model is thereby built. Numerical results show that various cost factors have significant effects on the Bitcoin futures price. We employ Monte Carlo simulation to approximate the Bitcoin futures price and we use Python to program the computations.
In 2008 a group of programmers, alias Satoshi Nakamoto, introduced bitcoin. Bitcoin is a cryptocurrency \nor virtual money derived from mathematical cryptography and is conceived as an alternative to government authorised \ncurrency. The founder anticipated, through bitcoin’s construction and his digital mining processes, that bitcoin prices \nwould be relatively stable. However, the recent bitcoin price decline proves that bitcoin is extraordinarily volatile and is \nnot that stable as hoped. Although some scientists have already shown that the fundamental value of bitcoin is zero, the \nprice of bitcoin has reached over 19.000$ in December 2018. Since then, bitcoin prices dropped nearly 70% from their \npeak value and showed in addition to that the typical trends of a speculative bubble. \nHyman Minsky and Charles Kindleberger discussed three different patterns of speculative bubbles. One is when price \nrises in an accelerating way and then crashes very sharply after reaching its peak. Another is when the price rises and is \nfollowed by a more similar decline after reaching its peak. The third is when the price rises to a peak, which is then \nfollowed by a period of gradual decline known as the period of financial distress, to be followed by a much sharper crash \nat some later time. One of the key findings of this study is that all these three patterns occurred during 2017-18 for the \nbitcoin price. \nTherefore, the purpose of this paper is to analyse the historical bitcoin prices in context with the typical five-step \ncharacteristics of a speculative bubble. Furthermore, each phase of a speculative bubble is explained by a behavioural \nfinance approach and answer the price development of this cryptocurrency. The result is frightening, bitcoin can be seen \nas a perfect textbook example of a speculative bubble.
This project uses software development to investigate the link between software and finance. The focus of the work is developing and implementing a trading algorithm which seeks to make profit by making trades based on arbitrage opportunities between currencies. Specifically, the sets of currencies examined are two fiat currencies and one cryptocurrency. Trades are made by combining a blockchain system, which maintains the cryptocurrency, and the live foreign exchange market, which enables fiat currency exchange. The main methodologies for carrying out the research are test-driven development and the use of a simulation to facilitate trades. By passing all of the unit tests, the software is verified. In addition, data gathered during runs of the simulation show that the algorithm successfully identifies arbitrage opportunities and turns a profit on average over many runs. This project proposes an interesting topic for further research in the field of blockchain technology used for financial trading.
Paulo VÃtor Jordão da Gama Silva, Augusto F.C. Neto, Marcelo Cabús Klötzle, Antônio Carlos Figueiredo Pinto · 5 authors
Research has shown that behavioral anomalies affect investors' choices and decisions in the financial markets. One such behavioral anomaly is feedback trading, a phenomenon wherein the investor uses past data to make future decisions. Using Sentana and Wadhwani's (1992) methodology, the 50 most liquid digital currencies (with the most extensive daily data reporting) were analyzed during the period 2013–2018. Results analysis suggests negative feedback trading in Tether Dollar and positive feedback trading in Bitcoin, Ethereum, CassinoCoin and ECC.