This paper investigates the relations between multiple measures of investor sentiment and the returns, volatility, trading volume, and liquidity. Using both data outside and inside market, we find that the Bullishness from socio-finance model are significant related to future realized volatility and trading volume, similar to Tweet, which is thought to capture information of well-informed investors in Bitcoin market
With the volume of activities associated with trading, it has become a very tedious task. The advent of the algorithmic trading has brought with it some positive change such as reduced latency and increase in liquidity in the Financial Market. The Algorithmic Trading also came with some high demands for the technological know-how and the resources to run it. This has put the retail trader in a seemingly disadvantaged position as these algo-programs are carefully guided secrets by those that have access to it. Crypto-currency can no longer be ignored as the concept is forming the bedrock for future transactions. Although highly publicized, the concept of these smart contracts is not really known. Looking into the future where the cryptocurrencies dominates over the traditional currencies, it has become imperative to give the individual trader/ retailer an additional tool to demystify the “black-Box” of the trading crypto-pairs with algorithmic trading strategy. The techniques employed are: Long Short-Term Memory (LSTM), Auto-regressive integrated moving average (ARIMA), Moving Average (MA), Cumulative Moving Average (CMA), and Artificial Neural Networks (ANN). The models performance will be measured via correlation, Mean Percentage Error (MPE), Percentage Error (MAPE), Mean Square Error (RMSE) standard deviation and Sharpe ratio (for the trading models).
Jan 1, 2020·Proceedings of the Proceedings of the 1st International Conference on Statistics and Analytics, ICSA 2019, 2-3 August 2019, Bogor, Indonesia
The purpose of these studies are to obtain bitcoin price predictions using three different approach in forecasting methods : ARIMA model, α-sutte indicator and LSTM algorithm, and to find out the accuracy level of the three methods in forecasting bitcoin’s price as well. Bitcoin closing’s price each
The aim of this study is to develop a reliable forecasting method for cryptocurrency namely XRP Ripple Cryptocurrency. The daily price of Ripple cryptocurrency collected from 1st October 2019 until 30th November 2019. This study implemented a forecasting method of simple moving average and the weighted moving average. The mean absolute percentage error for the simple moving average is 2.75%. Meanwhile, the mean absolute percentage error for weighted moving average is 2.25%. Therefore, the weighted moving average is more reliable forecasting method for predicting the price of Ripple cryptocurrency. The finding of this study helps investors to develop an investment portfolio with lower risk and higher returns.
Blockchain and cryptocurrencies have risen to popularity in the recent years to a great extent due to its increasing trading volumes and huge capitalization in the market. These cryptocurrencies are being used not only for trading but are being accepted for monetary transactions as well these days. As the prices fluctuate and return on investment increases investors, traders and general public are showing increased interest towards bitcoin and altcoins. This research focuses on implementing forecasting models that will return accurate price predictions for cryptocurrencies. Prices for Bitcoin, Ethereum and Litecoin are predicted using the traditional forecasting model for timeseries ARIMA, the Prophet Model and deep learning algorithm LSTM. The results of the three models were evaluated and the LSTM Model was found to outperform the Prophet as well as the ARIMA model.
The aim of this study is to determine whether successful predictions for cryptocurrencies such as Bitcoin can be obtained with different methods. The reason why Bitcoin prices (Bitcoin / $) are used in the study is that this cryptocurrency is still the most widely used cryptocurrency in the market, and the idea that it will successfully represent the overall state of the cryptocurrencies market. Financial market series may contain fluctuations for some reason, such as speculations. It also usually includes nonlinear changes. Such features lead to failures in obtaining forecasts for financial time series. In this study, with the GARCH model, one of the classicial time series models and LS -SVM method, a machine learning method, predictions of the Bitcoin price series were obtained, and model performances were compared. In the study, between January 01, 2017 and February 29, 2020, 1155 daily Bitcoin price series ( ) was used. In both models, the Bitcoin price series and the volatilities of this series were used, and external variables were not included in the models. For both models, forecasts were obtained for periods of 1 month, 2 months and 3 months. For GARCH and LS -SVM models, out of sample successful forecasting rates according to MAPE ratios were 98,0347% -95,3423% for 1 month; 97,9544% -96,1307% for 2 months and 98,1272% -91,4874% for 3 months, respectively. The GARCH model has provided more successful results for all three periods. The finding of the study is that the GARCH model can be used to obtain forecasts for the crypto price series.
The study examines the stability of Bitcoin price/returns volatility using an AR-GARCH model. The data for the study were the daily closing Bitcoin prices obtained from the bitcoin,com website for the study period 01/01/2013 - 31/12/2017.
A.G.D.J. Premarathne, Malka N. Halgamuge, Ruwani Samarakody, Ampalavanapillai Nirmalathas
Cryptocurrency has as of late pulled in extensive consideration in the fields of economics, cryptography, and computer science due to it is an encrypted digital currency, peer- to- peer virtual forex produced using codes, and it is much the same as another medium of the trade like real cash. This study mainly focuses to combine the Deep Learning with Data parallelism and Cloud Computing Machine learning engine as “hybrid architecture” to predict new Cryptocurrency prices by using historical Cryptocurrency data. The study has exploited 266,776 of Cryptocurrency prices values from the pilot experiment, and Deep Learning algorithm used for the price prediction. The four hybrid architecture models, namely, (i) standalone PC, (ii) Cloud computing without data parallelism (GPU-1), (iii) Cloud computing with data parallelism (GPU-4), and (iv) Cloud computing with data parallelism (GPU-8) introduced and utilized for the analysis. The performance of each model is evaluated using different performance evaluation parameters. Then, the efficiency of each model was compared using different batch sizes. An experimental result reveals that Cloud computing technology exposes new era by performing parallel computing in IoT to reduce computation time up to 90% of the Deep Learning algorithm-based Cryptocurrencies price prediction model and many other IoT applications such as character recognition, biomedical field, industrial automation, and natural disaster prediction.
Jan 1, 2020·Proceedings of the Proceedings of the 1st International Conference on Statistics and Analytics, ICSA 2019, 2-3 August 2019, Bogor, Indonesia
In recent years, Bitcoin has attracted a lot of attention because of its nature that supports encryption technology and monetary units. For traders, Bitcoin becomes a promising investment since its fluctuating prices potentially draw high profit (the higher the risk the higher the return). Unlike co
Ana Fernández Vilas, Rebeca P. Dı́az Redondo, Anton Lorenzo Garcia
There is a consensus about the good sensing characteristics of Twitter to mine and uncover knowledge in financial markets, being considered a relevant feeder for taking decisions about buying or holding stock shares and even for detecting stock manipulation. Although Twitter hashtags allow to aggregate topic-related content, a specific mechanism for financial information also exists: Cashtag (consisting of the company ticker preceded by $) is a supporting mechanism to track financial tweets referring to a company listed in a stock market. However, according to our experiments and due to the lack of conventions in cashtags usage, the irruption of cryptocurrencies has resulted in a significant degradation on the cashtag-based aggregation of posts. Unfortunately, Twitter' users may use homonym tickers to refer to cryptocurrencies and to companies in stock markets, which means that filtering by cashtag may result on both posts referring to stock companies and cryptocurrencies. This research proposes automated classifiers to distinguish conflicting cashtags and, so, their container tweets by analyzing the distinctive features of tweets referring to stock companies and cryptocurrencies. As experiment, this paper analyses the interference between cryptocurrencies and company tickers in the London Stock Exchange (LSE), specifically, companies in the main and alternative market indices FTSE-100 and AIM-100. Heuristic-based as well as supervised classifiers are proposed and their advantages and drawbacks, including their ability to self-adapt to Twitter usage changes, are discussed. The experiment confirms a significant distortion in collected data when colliding or homonym cashtags exist, i.e., the same $ acronym to refer to company tickers and cryptocurrencies. According to our results, the distinctive features of posts including cryptocurrencies or company tickers support accurate classification of colliding tweets (homonym cashtags) and Independent Models, as the most detached classifiers from training data, have the potential to be trans-applicability (in different stock markets) while retaining performance.
The objective of the study is to determine whether the Bitcoin forks have produced significant effects on the cryptocurrency market. The event study methodology is used in this paper in order to determine the statistical significance of the abnormal return of leading cryptocurrencies after three Bitcoin forks. The forks were viewed as three isolated events, with the estimations windows and the event windows constructed separately for each of them. There were statistically significant negative effects related to the creation of Bitcoin Gold and Bitcoin SV. Contrary to expectations, there was no statistically important effect throught out the most famous Bitcoin forking and emergence of Bitcoin Cash. Although cryptocurrencies are a current topic, the literature lacks quantitative research dealing with price changes. Without quantitative analysis, it is difficult to conclude whether the return change is a consequence of a statistically significant event The analysis would therefore provide the tool to determine the statistical significance of their impact on the market. A small number of observed cryptocurrencies is the main limitation of this research. Future researches could cover a wider scope of the market and include other famous cases of forking, for example, the Ethereum forks.
Romina Torres, Miguel A. Solís, Rodrigo Salas, Aurelio F. Bariviera
Cryptocurrencies have been receiving the sustained attention of investors since 2009. These new investment vehicles are digitally native, meaning that they are traded exclusively on 24/7 digital platforms. Consequently, they offer an excellent scenario to test the Efficient Market Hypothesis, by developing algorithm-based trading strategies. Such strategies aim to beat the market. It has been previously reported that daily returns do not exhibit long range dependence. However, daily volatility in major cryptocurrencies is highly persistent. Therefore, buy/hold/sell decision support systems could be able to capture such market inefficiency. This is especially important for investors interested in periodically trading a set of cryptocurrencies, in order to maximize their wealth. This paper presents a dynamic linguistic decision making approach for building decision models to support cryptocurrency investors in buy/hold/sell decisions. This approach exhibits a good computational performance for obtaining recommendations based on quantitative data. Moreover, this procedure is able to identify some inefficient cryptocurrency behaviors which are not captured by traditional econometric techniques. Our results uncover arbitrage opportunities that outperform buy-and-hold or random strategies.
Xinwen Ni, Taojun Xie, Wolfgang Karl Härdle, Xiaorui Zuo
Abstract Cryptocurrency markets are highly sensitive to regulatory changes, often experiencing sharp price fluctuations in response to new policies and government interventions. Despite this, existing market indices fail to adequately capture the risks associated with regulatory uncertainty. In this paper, we introduce the Cryptocurrency Regulatory Risk Index (CRRIX), a machine learning-based index designed to quantify the impact of regulatory developments on cryptocurrency markets. Our methodology employs Latent Dirichlet Allocation (LDA) to classify policy-related news articles from major cryptocurrency news platforms, providing an objective measure of regulatory risk. We find that the CRRIX exhibits strong synchronicity with VCRIX, a cryptocurrency volatility index, suggesting that regulatory uncertainty plays a significant role in driving market fluctuations. Our results indicate that regulatory risk is a leading factor in market volatility, with major policy shifts triggering significant market movements. The proposed regulatory risk index provides a novel approach to quantifying policy uncertainty in the cryptocurrency sector, offering valuable insights for market participants navigating this rapidly changing environment.
M. Sivaram, E. Laxmi Lydia, Irina V. Pustokhina, Denis A. Pustokhin · 7 authors
The booming applications of bitcoin Blockchain technologies made investors concerned about the return and risk of financial products. So, the return rate of bitcoin must be foreseen in prior. This research article devises an effective return rate prediction technique for Blockchain financial products based on Optimal Least Square Support Vector Machine (OLS-SVM) model. The parameter optimization of the LS-SVM model was performed using hybridization of Grey Wolf Optimization (GWO) with Differential Evolution (DE), called optimal GWO (OGWO) algorithm. The hybridization process is performed to eliminate the local optima problem of GWO and enhance the diversity of the population. To verify the goodness of the proposed model, the Ethereum (ETH) return rate was chosen as the target and experimental analysis was performed on it to verify the predictive results on the time series. The experimental outcome was analyzed in terms of two performance measures namely Mean Squared Error (MSE) and Mean Absolute Percentage Error (MAPE). The obtained simulation outcome infers that the OLS-SVM model yielded better predictive outcome of the return rate of financial products.
The application of machine learning algorithms in predicting cryptocurrency prices has gained significant attention in recent years. Researchers have explored various approaches such as recurrent neural networks, deep learning neural networks, Bayesian regression, k-nearest neighbor, support vector machine, and other algorithms to forecast the prices of cryptocurrencies like Bitcoin, Ethereum, Dogecoin and Litecoin. This paper will draw on established literature on price prediction using machine learning, including studies on NFT sales predictability, NFT sale price fluctuations prediction, gold price prediction, and silver price forecasting. The research paper has focused on utilizing high-dimensional features, time-series analysis, as well as the comparison of different statistical models and machine learning algorithms. Additionally, the prediction models have incorporated factors such as market liquidity, exchange market dynamics. While the literature acknowledges the potential of machine learning in cryptocurrency price prediction, gold, silver and NFT’s there is a recognized gap in the application of these techniques across a broader range of cryptocurrencies. The proposed methodology will integrate various machine learning models and statistical methods to predict the prices of cryptocurrencies, gold, silver, and NFTs, taking into account factors such as market trends, trade networks and visual features. Furthermore, the studies emphasize the importance of feature engineering, sample dimension engineering, and the use of various machine learning techniques to enhance the accuracy and stability of cryptocurrency price predictions. As the cryptocurrency market continues to expand, there is a need for further research to develop robust machine learning models that can effectively forecast the prices of diverse cryptocurrencies, contributing to the advancement of this field.
Bitcoin is electronic money that can be used as an alternative for investment. Investors will get benefit buying bitcoin when the price of bitcoin is down and reselling it when bitcoin prices are increasing. The fluctuating bitcoin prices cause forecasting as a basis for investors to make decisions, where the time series method is used as a forecasting model, then a pattern can be found to predict future events. The classical time series methods are often violating the statistical assumptions. To face these problems, then it is used free assumptions methods, the method with the Fuzzy Time Series Markov Chain, the Chen Logical Method, and its segmented methods due to unbalancing forecasting results. This study is built the forecasting model of the price of bitcoin for the coming period based on the data from 2010 to 2020. The proposed methods have a better fit for bitcoin time series data prices. Besides, the Fuzzy Time Series Markov Chain method has the slightly smallest accuracy error based on Mean Absolute Percentage Error (MAPE) comparing to the Fuzzy Time Series Segmented Chen Logical Method and Fuzzy Time Series Chen Logical Method.
Johannes Jakubik, Abdolreza Nazemi, Andreas Geyer-Schulz, Frank J. Fabozzi
In this paper, we investigate how a deep learning machine learning model can be applied to improve Bitcoin price forecasting and trading by incorporating unstructured information from financial news. The two-stage model we propose that includes financial news significantly outperforms machine learning models without financial news. In the first stage, we leverage long short-term memory (LSTM) networks to extract structured information from financial news. In the second stage, we apply machine learning models with structured input from financial news to the prediction of Bitcoin prices. In addition to the superior performance relative to machine learning models without input from financial news, we find that the out-of-time rate of return attained with the proposed forecasting system is substantially higher than for a buy-and-hold strategy. Our study highlights how combining deep learning and financial news offers investors and traders support for the monetization of unstructured data in finance.
This paper investigates the risk and return properties of a trading strategy for the cryptocurrency market. The main predictive power for portfolio formation comes from a simple prospect theory model that only uses price information readily available. The dataset consists of a large body of cryptocurrencies from 2014 to 2020. I find a strong outperformance over the market, even after controlling for known predictors. Factor regressions with a cryptocurrency three-factor model further reveal significant alphas. Robustness test emphasize the legitimacy of the strategy. On average, cryptocurrencies with a high (low) prospect theory value earn low (high) subsequent returns. Interestingly, traders in the cryptocurrency market seem to assess the attractiveness of cryptocurrency in a way described by prospect theory. Mechanical tests of the model show that probability weighting is a main driver behind this assessment. Cryptocurrencies with a high prospect theory value tend to be highly positively skewed. This skewness could be the reason why the cryptocurrency seems attractive to traders, similar to lottery-like gambles.