Cryptocurrencies are a type of digital money distinguished by a decentralized system that uses encryption to authenticate transactions and keep records, obviating the need for a central authority. A key element of these digital assets is the decentralization of power from a single entity to a dispersed network. The extreme price volatility of cryptocurrencies, on the other hand, has a significant influence on international commerce, making precise price forecasting critical for investors and traders. In this study, the investigation is made on how deep learning models, especially the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU), may be used to solve the problem of excessive price swings in cryptocurrencies like as Bitcoin, Ethereum, Litecoin, and Dogecoin. Previous research has looked at forecasting methods such as ARIMA and Support Vector Machines (SVM), but the findings have fallen short of the promising results obtained by the deep learning models studied in this work. Our major objective is to create strong forecasting models based on LSTM and GRU, with an emphasis on their ability to make credible forecasts for bitcoin values. We find that GRU consistently beats LSTM for the majority of the cryptocurrencies under consideration by comparing their performance using two separate error prediction approaches, namely mean absolute percentage error (MAPE) and root mean square error (RMSE). This study's findings help to enhance prediction methods for understanding and reducing the impact of cryptocurrency price volatility on international commerce. Deep learning algorithms for projecting bitcoin values give significant insights for investors and decision-makers navigating the volatile and ever-changing terrain of the crypto market.
Since cryptocurrencies are becoming more widely used and accepted in the financial system, precise price forecasting is essential for optimizing bitcoin investments. In this research study, we evaluated various machine learning models, including linear regression (LR), decision tree regression (DT), random forest regression (RF), support vector regression (SVR), gradient boosting regression (GB), adaboost regression, extreme gradient boosting regression (XGR), light gradientboosting regression (LGBM), k-nearest neighbors regression (KNN), ridge, andlasso. Additionally, we incorporated two deep learning (DL) models, namely artificial neural networks (ANN) and convolutional neural networks (CNN), to forecast daily bitcoin prices (BP). The initial data was obtained from Kaggle, a well-known platform for data science projects, and we applied the min-max scaler technique for consistent scaling during preprocessing. To assess the predictive capabilities of the models, we utilized regression metrics such as root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (R). Based on our findings, the CNN model demonstrated the highest effectiveness in predicting BPs among the DL models, with an RMSE of 0.0543, MAE of 0.0324, and an R value of 0.960. In the case of machine learning models, the RF model outperformed others, achieving an RMSE of 0.0246 and MAE of 0.0561. Investors, scholars, and decision-makers may all gain from these findings’ insightful revelations about BP forecasting. Developing these models further, investigating different preprocessing methods, and expanding the analysis to other cryptocurrencies might be the main goals of future research.
Bitcoin is always interesting to keep predicting where the next price movement will go. Bitcoin is the first and most influential Cryptocurrency on cryptocurrency price movements. Bitcoin is traded in many markets, the largest in Indonesia is Indodax. Indodax provides a document sharing API so that third parties can build applications that are able to process data on bitcoin price movements in real-time and continuously. This research shows how patterned datasets can be applied to monitor bitcoin price movements from the indodax market and show their effects on other cryptocurrency assets besides bitcoin. This research shows how data that is patterned and then processed using the minimum and maximum functions can provide 2 important information, namely the potential position of bitcoin when it is at the maximum and minimum points. The results of the patterned dataset formula are then compared to the movements of the 2 cryptocurrencies with the largest capitalization, namely BTC (Bitcoin) and ETH (Ethereum). The results of the comparison show that the use of patterned dataset formulas successfully shows important points in cryptocurrency trading, with simpler instructions.
In the realm of cryptocurrency forecasting, accurately predicting short-term Bitcoin log returns remains a challenging endeavor due to its inherent volatility and sensitivity to multifarious external factors. This study addresses this challenge by proposing an integrated approach that combines the capabilities of the TimesNet deep learning model with sentiment analysis techniques. TimesNet, specifically designed for time series data, has demonstrated proficiency in extracting salient patterns. When synergized with sentiment analysis, a more nuanced understanding of price determinants emerges. Preliminary results from our experiments indicate a significant enhancement in predictive accuracy within the Bitcoin market. Such advancements not only furnish investors and researchers with refined forecasting tools but also accentuate the burgeoning role of deep learning methodologies in the domain of financial forecasting.
Accurate cryptocurrency price prediction is essential to investors and researchers for analyzing trends and advising financial decisions, as price prediction is fundamental to making beneficial investment decisions. Due to the high volatility and unpredictability of the cryptocurrency market, it is difficult to predict these prices based on cryptocurrency time series data accurately. This research paper presents a two-fold analysis of the effectiveness of neural networks and deep learning to predict cryptocurrency prices and proposes a novel approach to cryptocurrency price prediction. This is done by considering Long-Short Term Memory (LSTM) and Transformer neural networks that use historical price features in addition to volatility and momentum technical indicators, along with historical price features, and testing these models on Bitcoin (BTC), Ethereum (ETH) and Litecoin (LTC). Momentum and volatility technical indicators such as Relative Strength Index (RSI), Bollinger Bands %B and Moving Average Convergence/Divergence (MACD) are not commonly used in cryptocurrency machine learning models. Still, the addition of these features can give better insight into the general trend of the price. By adding volatility and momentum features to our LSTM and Transformer models, we see a significant increase in price prediction accuracy, and we also find that Transformers tend to outperform LSTM models in price prediction and trends of cryptocurrency data.
Hardik Choudhary, M. J. Shukla, S Raghavendra, Ramyashree Ramyashree
The development of the fintech industry has transformed cryptocurrencies into intangible assets and opened many opportunities in the fields of financial research and quantitative markets. Cryptocurrencies are a type of electronic currency used to conduct transactions in the financial system. In addition to the technical analysis that a trader typically does, it has been established over time that market mood is extremely important in determining market conditions. This document provides a method for estimating cryptocurrency prices based on historical data and user sentiment. To achieve this, a long short-term memory (LSTM) model and sentiment analysis of tweets were used. Furthermore, it was supported by the outcomes, as the LSTM model demonstrated a precision of 69.32%, which is respectable when it comes to the forecasting of financially risky assets like bitcoin. The final accuracy attained was 70%, indicating that the model will accurately recommend buying or selling in about 3 out of every 4 scenarios that it is presented with. Traders can achieve a high alpha with a risk reward ration of 1:2 to benefit from this research finding and can combine the findings with technical indicators to produce better trades. This research has a very large application in the field of quant trading. Findings in this research can be used to build multiple models with multiple attributes which will improve the overall accuracy and precision of trades.
Mihailo Todorović, Aleksandar Petrović, Ana Toskovic, Miodrag Živković · 6 authors
This study focuses on analyzing historical data to forecast future trends in Bitcoin prices due to its influence on the business landscape. Its high volatility attracted attention to understanding the influencing factors for its price. This paper presents an empirical investigation using time-series data of various exogenous and endogenous variables. Closing prices of Bitcoin and Ethereum, along with the daily volume of Bitcoin-related tweets are examined for Bitcoin closing price prediction by a long-short term memory (LSTM) network, fine-tuned by a hybrid adaptive reptile search algorithm. The analysis covers a three-year period, in which data is divided into training, validation, and testing sets. Comparative analysis against LSTM networks tuned by other high-performing metaheuristic algorithms demonstrates that the novel approach outperforms competitors in terms of standard regression metrics.
El mercado de criptomonedas está experimentando un rápido crecimiento, lo que lo convierte en una alternativa potencialmente más lucrativa que los mercados financieros convencionales. No obstante, esta expansión va de la mano con una significativa volatilidad, presentando asà un desafÃo crucial. En el contexto de esta tesis de maestrÃa, se desarrollaron modelos de predicción de series temporales para el precio de cierre de Bitcoin mediante el uso de algoritmos de aprendizaje profundo, tales como LSTM y GRU. Además, se llevó a cabo una comparación con modelos tradicionales como ARIMA, con el propósito de analizar y evaluar su rendimiento.
Jinghua Wang, Geoffrey Ngene, Yan Shi, Ann Nduati Mungai
Policymakers and portfolio managers pay keen attention to sources of uncertainties that drive asset returns and volatility. The influence of uncertainty on Bitcoin has the potential to drive fluctuations in the entire cryptocurrency market. We investigate the predictability of thirteen economic policy uncertainty indices on Bitcoin returns. Using the Random Forest machine learning algorithm, we find that Singapore’s economic policy uncertainty (EPU) has the strongest predictive power on Bitcoin returns, followed by financial crisis (FC) uncertainty and world trade uncertainty (WTU). We further categorize these uncertainties into different groups. Interestingly, the predictability of uncertainty indices on Bitcoin returns within the international trade group is stronger compared to other uncertainty categories. Additionally, we observed that internet-based uncertainty measures have more predictive power of Bitcoin returns than newspaper- and report-based measures. These results are robust using various additional machine learning methods. We believe that these findings could be valuable for policymakers and portfolio managers when making decisions related to uncertainty drivers of cryptocurrency prices and returns.
As an alternate form of trade money, cryptocurrencies are now widely accepted and have been integrated into all financial activities. Trading cryptocurrencies is one of the most popular and promising forms of investing. However, the cryptocurrency markets are notorious for their tremendous volatility and price discrepancies over short periods. To ensure accurate and trustworthy predictions, it is necessary to have an automated trading model that includes portfolio management and optimization. Automated algorithmic trading uses computers to carry out trades in accordance with the past and predicted trends following a defined set of rules. These rules include trading instructions based on time, value, quantity, or any other mathematical model of trading. Profits may be achieved through algorithmic trading at inhumanely high speeds and frequencies. In addition to providing profitable trading opportunities, algorithmic trading increases market fluidity and increases trading accuracy by reducing human elements such as emotions and feelings regarding the trade. This paper aims to contribute to the ongoing market revolution by discussing the various aspects of cryptocurrecy trading, its forecasting and actual development of an algorithmic trading bot that will implement client strategies closely accompanied by its own calculations for daily exchanges based on economic conditions and client approaches. It will also contribute to and exchange with ongoing adjustments throughout the day to ensure the best profitability of the clients.
Bitcoin is one of the cryptocurrencies that have a large number of holders and a high transaction volume. Its price has experienced major fluctuation since its inception, resulting in relatively high price volatility. This has been a challenge for researchers over the past decade in forecasting its future price. In this study, we aim to address this problem by utilizing a neural network model called the Temporal Fusion Transformer (TFT). The TFT model uses the concept of multi-horizon forecasting to learn from historical data and predict future prices based on multiple time steps. Due to the numerous factors influencing Bitcoin's price, our study incorporates additional predictive data, such as Twitter sentiment, trends, and seasonality. The proposed method is trained using the TFT model with historical Bitcoin data from 2014 until 2020, then we performed transfer learning by using different past covariates as inputs. As a result, the proposed method in this study demonstrated the best performance, with a MAE of 0.05, a RMSE of 0.07, a MAPE of 5.92%, and a quantile loss of 0.03, surpassing other approaches. This makes it the state-of-the-art solution for highly volatile forecasting using a neural network model.
In recent years, cryptocurrencies have become more and more popular and recognized, and they have the characteristics of financial products and are involved in many financial transactions. Therefore, cryptocurrency trading is generally considered one of the most popular and promising types of profitability. However, this growing financial market is unstable, so this paper aims to develop a relatively accurate and reliable prediction model. This paper compares and analyzes various neural networks and traditional analytical methods based on the need to forecast cryptocurrency prices. Finally, this paper uses a variety of deep learning neural networks, including Recurrent neural network (RNN), Long Short-Term Memory (LSTM) network, Gate Recurrent Unit (GRU), Bidirectional Long Short-Term Memory (BiLSTM), and Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM). The data of three cryptocurrencies with the highest market capitalization, largest size, and best known were selected, namely Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB). We collect the three-cryptocurrency data for five consecutive years. The critical data collected includes daily (Eastern Standard Time) open, close, low, and high prices and the market capacity. The findings show that the CNN-LSTM neural network model has some limitations, such as the time lag in prediction. However, it can predict prices more accurately than other deep learning methods and have an overwhelming advantage over traditional time series methods.
Abstract Cryptocurrencies and Bitcoin, in particular, are prone to wild swings resulting in frequent jumps in prices, making them historically popular for traders to speculate. It is claimed in recent literature that Bitcoin price is influenced by sentiment about the Bitcoin system. Transaction, as well as the popularity, have shown positive evidence as potential drivers of Bitcoin price. This study introduces a bivariate jump-diffusion model to capture the dynamics of Bitcoin prices and the Bitcoin sentiment indicator, integrating trading volumes or Google search trends with Bitcoin price movements. We derive a closed-form solution for the Bitcoin price and the associated Black–Scholes equation for Bitcoin option valuation. The resulting partial differential equation for Bitcoin options is solved using an artificial neural network, and the model is validated with data from highly volatile stocks. We further test the model’s robustness across a broad spectrum of parameters, comparing the results to those obtained through Monte Carlo simulations. Our findings demonstrate the model’s practical significance in accurately predicting Bitcoin price movements and option values, providing a reliable tool for traders, analysts, and risk managers in the cryptocurrency market.
Rasoul Amirzadeh, Dhananjay Thiruvady, Asef Nazari, Mong Shan Ee
Despite advances in artificial intelligence-enhanced trading methods, developing a profitable automated trading system remains challenging in the rapidly evolving cryptocurrency market. This research focuses on developing a reinforcement learning (RL) framework to tackle the complexities of trading five prominent altcoins: Binance Coin, Ethereum, Litecoin, Ripple, and Tether. To this end, we present the CausalReinforceNet~(CRN) framework, which integrates both Bayesian and dynamic Bayesian network techniques to empower the RL agent in trade decision-making. We develop two agents using the framework based on distinct RL algorithms to analyse performance compared to the Buy-and-Hold benchmark strategy and a baseline RL model. The results indicate that our framework surpasses both models in profitability, highlighting CRN's consistent superiority, although the level of effectiveness varies across different cryptocurrencies.
We provide a comprehensive investigation into the profitability of technical trading methods applied to the cryptocurrency pairs BTC/USDT and ETH/USDT. By employing rigorous evaluations and incremental examinations, we address the pervasive issue of data-snooping bias that often plagues the evaluation of trading strategies. Our empirical results indicate the lack of profitable technical trading strategies in both the analysis sample and prediction sample periods, even after rigorous adjustments for data snooping. These findings highlight the difficulties associated with selecting profitable technical trading strategies in the dynamic and volatile cryptocurrency market. Market participants, including individual traders, institutional investors, and regulatory bodies, should take note of our findings when making investment decisions based on technical analysis.
Automated Market Makers (AMMs) are major centers of matching liquidity supply and demand in Decentralized Finance. Their functioning relies primarily on the presence of liquidity providers (LPs) incentivized to invest their assets into a liquidity pool. However, the prices at which a pooled asset is traded is often more stale than the prices on centralized and more liquid exchanges. This leads to the LPs suffering losses to arbitrage. This problem is addressed by adapting market prices to trader behavior, captured via the classical market microstructure model of Glosten and Milgrom. In this paper, we propose the first optimal Bayesian and the first model-free data-driven algorithm to optimally track the external price of the asset. The notion of optimality that we use enforces a zero-profit condition on the prices of the market maker, hence the name ZeroSwap. This ensures that the market maker balances losses to informed traders with profits from noise traders. The key property of our approach is the ability to estimate the external market price without the need for price oracles or loss oracles. Our theoretical guarantees on the performance of both these algorithms, ensuring the stability and convergence of their price recommendations, are of independent interest in the theory of reinforcement learning. We empirically demonstrate the robustness of our algorithms to changing market conditions.
Omar M. Ahmed, Lailan M. Haji, Ayah M. Ahmed, Nashwan M. Salih
The field of finance makes extensive use of real-time prediction of stock price tools, which are instruments that are put to use in the process of creating predictions. In this article, we attempt to predict the price of Bitcoin in a manner that is both accurate and reliable. Deep learning models, as opposed to more traditional methods, are used to manage enormous volumes of data and to generate predictions. The purpose of this research is to develop a method for predicting stock prices using the Hybrid Convolutional Recurrent Model (HCRM) architecture. This model architecture integrates the advantages of two separate deep learning models: The 1-Dimensional-Convolusional Neural Network (1D-CNN) and the Long-Short Term Memory (LSTM). The 1D-CNN is responsible for the feature extraction, while the LSTM is in charge of the temporal regression. The developed 1D-CNN-LSTM model has an outstanding performance in predicting stock values.
Purpose This paper aims to investigate the role of price-based information from major cryptocurrencies, foreign exchange, equity markets and key commodities in predicting the next-minute Bitcoin (BTC) price. This study answers the following research questions: What is the best sparse regression model to predict the next-minute price of BTC? What are the key drivers of the BTC price in high-frequency trading? Design/methodology/approach Least absolute shrinkage and selection operator and Ridge regressions are adopted using minute-based open-high-low-close prices, volume and trade count for eight major cryptos, global stock market indices, foreign currency pairs, crude oil and gold price information for February 2020–March 2021. This study also examines whether there was any significant break and how the accuracy of the selected models was impacted. Findings Findings suggest that Ridge regression is the most effective model for predicting next-minute BTC prices based on BTC-related covariates such as BTC-open, BTC-high and BTC-low, with a moderate amount of regularization. While BTC-based covariates BTC-open and BTC-low were most significant in predicting BTC closing prices during stable periods, BTC-open and BTC-high were most important during volatile periods. Overall findings suggest that BTC’s price information is the most helpful to predict its next-minute closing price after considering various other asset classes’ price information. Originality/value To the best of the authors’ knowledge, this is the first paper to identify the covariates of major cryptocurrencies and predict the next-minute BTC crypto price, with a focus on both crypto-asset and cross-market information.
Cryptocurrency is a digital or virtual currency secured by cryptography. Using cryptography, this digital currency is nearly impossible to counterfeit. All transaction records are stored on the blockchain. Ethereum is one of the cryptocurrencies, along with Bitcoin, etc. Ethereum was first created in 2013 by Vitalik Buterin, a professional programmer, with the goal of building decentralized applications. Ethereum was launched in 2015 with its own cryptocurrency called ether. Proof of Stake (PoS) is the underlying mechanism that activates validators after receiving sufficient stake. For Ethereum, users need to stake 32 ETH to become validators. The objectives of this research are to understand the general overview of ETH prices in the past two years, assess the effectiveness of transitioning from Proof of Work (PoW) to PoS, and explore the application of long-short-term memory (LSTM) in price prediction. The results of this research indicate that the LSTM model is well-suited for predicting prices with a sufficiently large and complex dataset. The evaluation parameters, including the root mean square error (RMSE) with a value of 130.109, the mean squared error (MSE) with a value of 16928.408, and the mean absolute error (MAE) with a value of 92.9926, show that the performance of the LSTM model is good.
In recent years, cryptocurrencies have gained a lot of popularity in the financial markets and now, in addition to investing on them, it is possible to use them as a common currency to meet daily needs. Given the complex nature of financial markets and their reliance on different parameters to determine stocks' and assets' prices, the ability to predict prices is important for investment decisions, especially with respect to cryptocurrencies. To this end, Deep Learning (DL)-based algorithms can be viable solutions, owing to their use as time series forecasting tools. In this paper, we investigate the applicability of DL algorithms to forecast the prices of three cryptocurrencies, namely Bitcoin, Ethereum, and Ripple. We evaluate the performance of the proposed approach, in terms of short-term and long-term prediction accuracy (considering proper error metrics).
The accurate prediction of an uncertain future is crucial for society. Particularly when it comes to finance, where it holds vital importance for companies. Time series analysis is conducted to forecast the potential value of financial assets in the near future. Due to the impracticality of manually analyzing hundreds of indicators, machine learning methods are frequently employed. This study showcases the optimal value of the lag hyperparameter used in Random Forest and Extreme Gradient Boosting methods. The daily Bitcoin OHLCV data of a time series was enriched with indicators, and it was observed that the lag hyperparameter yielded the best results when considering the preceding 23 time series units.
Market traders often earn income by buying and selling assets through market transactions with the goal of maximizing total returns. Since each asset purchased has a certain amount of gain and loss, traders need to develop a trading strategy, which may affect their total profit. In the case of gold and bitcoin, this paper develops an ARIMA daily price forecasting model and a trade strategy model based on price increases to help market traders develop optimal trading strategies for gold and bitcoin.