This study evaluates the effectiveness of the CNN-LSTM hybrid model in predicting the Ethereum exchange rate against the United States Dollar (USD) by comparing the performance of the model without optimization and the model with hyperparameter optimization using Bayesian Optimization. The dataset used is sourced from Yahoo Finance covering the period 2017-2023. The results show that the CNN-LSTM model with hyperparameter optimization consistently outperforms the model without optimization, with improved prediction accuracy shown through the RMSE, MAE, MAPE, and R² values. Hyperparameter optimization resulted in an optimal configuration with 166 filters, kernel size 5, 168 LSTM units, 91 dense units, learning rate 0.00114, and batch size 32. This research confirms the effectiveness of the CNN-LSTM hybrid approach in predicting crypto exchange rates, and demonstrates the importance of hyperparameter optimization in improving prediction accuracy.
Cryptocurrencies have transformed financial markets with their innovative blockchain technology and volatile price movements, presenting both challenges and opportunities for predictive analytics. Ethereum, being one of the leading cryptocurrencies, has experienced significant market fluctuations, making its price prediction an attractive yet complex problem. This paper presents a comprehensive study on the effectiveness of Large Language Models (LLMs) in predicting Ethereum prices for short-term and few-shot forecasting scenarios. The main challenge in training models for time series analysis is the lack of data. We address this by leveraging a novel approach that adapts existing pre-trained LLMs on natural language or images from billions of tokens to the unique characteristics of Ethereum price time series data. Through thorough experimentation and comparison with traditional and contemporary models, our results demonstrate that selectively freezing certain layers of pre-trained LLMs achieves state-of-the-art performance in this domain. This approach consistently surpasses benchmarks across multiple metrics, including Mean Squared Error (MSE), Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE), demonstrating its effectiveness and robustness. Our research not only contributes to the existing body of knowledge on LLMs but also provides practical insights in the cryptocurrency prediction domain. The adaptability of pre-trained LLMs to handle the nature of Ethereum prices suggests a promising direction for future research, potentially including the integration of sentiment analysis to further refine forecasting accuracy.
Md. Shahidul Islam, Monjira Bashir, Siddikur Rahman, Md Abdullah Al Montaser ¡ 7 authors
The cryptocurrency market, with its record volatility and breakneck speed, is a revolutionary phenomenon that is reshaping the entire world's landscape. Unlike regular markets, cryptocurrencies undergo unprecedented volatility caused by a complex interaction of factors ranging from speculative trading to updates in regulations, technological innovations, and macroeconomic trends. The central objective of this research was to develop and evaluate machine learning-driven models of cryptocurrency price trend forecasting. The focus of this research project revolved around prominent cryptocurrencies, i.e., Bitcoin (BTC), Ethereum (ETH), and other prominent altcoins, within the United States. The dataset employed in this analysis comprises vast historical price data, trading volumes, and key market indicators of major cryptocurrencies, i.e., Bitcoin (BTC), Ethereum (ETH), and other major altcoins. Historical price data is presented in terms of daily, hourly, and minute-level opening, closing, high, and low prices, providing detailed insights into temporal price behavior. Trading volumes, which reflect the intensity of trading action, are also provided to represent liquidity and investor participation behavior. The dataset also includes various market indicators, i.e., moving averages, relative strength index (RSI), Bollinger Bands, and other technical indicators, which play a pivotal role in establishing market patterns and momentum. Three models are chosen in this study: Logistic Regression, Random Forest Classifier, and XG Boost Classifier. For classification models, accuracy, precision, recall, and F1-score metrics are employed to evaluate the performance of the models in terms of predicting the directions of the markets (e.g., upward or downward directions). With the highest accuracy, Logistic Regression was the best-performing of the models tested, showing its relative superiority. The integration of AI forecasts into cryptocurrency trading has the potential to revolutionize the United States financial markets by providing traders and institutional investors with advanced tools to make decisions. The use of AI tools in cryptocurrency trading also has significant implications for United States regulation compliance. The integration of machine learning tools within cryptocurrency trading platforms is a significant step towards unleashing the true potential of AI in the financial markets. The field of AI-based cryptocurrency forecasting offers numerous areas of future research with the potential to break through present limitations and unlock new paths of market analysis. One of those areas is the use of deep learning models, i.e., Long Short-Term Memory (LSTM) networks, for time-series cryptocurrency forecasting.
Abstract Forecasting cryptocurrencies as a financial issue is crucial as it provides investors with possible financial benefits. A slight improvement in forecasting performance can lead to increased profitability; Therefore, obtaining a realistic forecast is very important for investors. Bitcoin, frequently mentioned in recent due to its volatility and chaotic behavior, has become an investment tool, especially during and after the COVID-19 pandemic. In this study, selected ML techniques were investigated for predicting cryptocurrency movements by using technical indicator-based data sets and measuring the applicability of the techniques to cryptocurrencies that do not have sufficient historical data. In order to measure the effect of data size, Bitcoinâs last 1 year and 7 years of data were used. Following the related literature, Google trends and the number of tweets were used as input features, in addition to the most commonly used twelve technical indicators. Random Forest, K-Nearest Neighbors, Extreme Gradient Boosting (XGBoost-XGB), Support Vector Machine (SVM), Naive Bayes (NB), Artificial Neural Networks (ANN), and Long-Short-Term Memory (LSTM) network were optimized for best results. Accuracy, F1, and area under the ROC curve values were used to compare the model performance. For continuous data, ANN and SVM performed the best with the highest accuracy and outperformed the other ML models for complete and reduced sets. LSTM reached the best accuracy for trend data, but SVM, NB, and XGB models showed similar performance. The research shows that some indicators significantly affect prediction performance, and the data discretization process also improved the modelâs accuracy. While the number of samples affects the results of many ML models, correctly optimized and fine-tuned models may also give excellent results even with less data.
Purpose Set against a rapidly evolving technologically driven investment landscape, this research aims to explore the complex interrelations among artificial intelligence, alternative energy stocks, eco-friendly investments, geopolitical risks (GPRs) and Ethereumâs energy consumption. Design/methodology/approach This work encompasses deploying the H2O Automated Machine Learning approach, explicitly focusing on analyzing market indicators. Additionally, the research emphasizes the evaluation of feature significance, identifying crucial variables that significantly influence the predictive outcomes. Besides, this study employs Shapley Additive Explanations to interpret the modelâs output, offering a detailed analysis of feature contributions and enhancing the modelâs transparency. Findings Key variables such as GPR, clean market (PBW) and the natural gas index (NG) significantly influence oil price predictions. The model demonstrates reliability, with areas for improvement in capturing unexplained variance. Practical implications This study offers valuable insights for energy sector market analysts, traders and policymakers, aiding in strategic decision-making and understanding market trends. Social implications This research emphasizes fostering clean and sustainable energy markets. It emphasizes the crucial role of advancements in artificial intelligence and renewable energy investments in accelerating the transition to environmentally responsible energy markets, highlighting their significance in fostering sustainability and mitigating climate change impacts. Originality/value This study pioneers integrating cutting-edge machine learning methodologies with crude oil market analysis, shedding light on critical influencing factors and forecasting aspects.
The aim of this paper is to forecast the volatility of Ethereum for a specified time period. To achieve this, we evaluated and compared the performance of five different models: Generalised Autoregressive Conditional Heteroscedasticity (GARCH), Long Short-Term Memory (LSTM), Random Walk Model (RWM), Neural Network Baseline Metrics - Fully Connected Network, and Baseline Model. After applying these models, we compared the estimated volatilities with the realized volatilities to assess the prediction accuracy. Considering the result comprehensively, the GARCH(1,1) is the most accurate model among these five. Our results revealed interesting insights into the nature of Ethereum, which behaves differently from traditional currencies. However, given Ethereum's early-stage behavior, future results may vary.
This paper presents an advanced framework for analyzing cryptocurrency market microstructure through the integration of deep learning techniques and social media sentiment analysis. The proposed approach combines BERT-based sentiment analysis with market microstructure indicators to capture complex market dynamics. The framework processes multi-source data streams, including social media content and order book information, to generate comprehensive market insights. Experimental evaluation conducted on cryptocurrency market data from January 2022 to December 2023 demonstrates superior performance compared to traditional approaches. The model achieves 91.2% prediction accuracy and maintains a Sharpe ratio of 2.34 in trading simulations. The attention mechanism effectively identifies relevant market signals with 92.3% precision, while the temporal feature extraction module captures multi-scale market patterns. The applications have been successful with the capability of the ability to below 100 milliseconds, fit for high applications. The studies made for fields by creating the processing system for market microstructure focuses for commercial and investigators. The framework's performance stability across different market conditions validates its practical applicability in cryptocurrency trading and market analysis.
The stock market is a vital component of the financial sector. Due to the inherent uncertainty and volatility of the stock market, stock price prediction has always been both intriguing and challenging. To improve the accuracy of stock predictions, we construct a model that integrates investor sentiment with Long Short-Term Memory (LSTM) networks. By extracting sentiment data from the âFinancial Postâ and quantifying it with the Vader sentiment lexicon, we add a sentiment index to improve stock price forecasting. We combine sentiment factors with traditional trading indicators, making predictions more accurate. Furthermore, we deploy our system on the blockchain to enhance data security, reduce the risk of malicious attacks, and improve system robustness. This integration of sentiment analysis and blockchain offers a novel approach to stock market predictions, providing secure and reliable decision support for investors and financial institutions. We deploy our system and demonstrate that our system is both efficient and practical. For 312 bytes of stock data, we achieve a latency of 434.42 ms with one node and 565.69 ms with five nodes. For 1700 bytes of sentiment data, we achieve a latency of 1405.25 ms with one node and 1750.25 ms with five nodes.
The integration of Artificial Intelligence (AI) in finance has significantly transformed various aspects of the industry, from algorithmic trading and risk management to regulatory compliance and decentralized finance (DeFi). AI-driven models enhance market prediction accuracy, automate trading strategies, and improve fraud detection, thereby increasing efficiency and reducing financial risks. Moreover, AI-powered robo-advisors and credit scoring systems contribute to financial inclusion by offering personalized and data-driven services. Despite these advancements, challenges such as AI explainability, data privacy concerns, algorithmic bias, and regulatory constraints remain critical research areas. Additionally, emerging trends, including quantum computing, AI-enhanced DeFi, and privacy-preserving machine learning, are expected to further shape the future of AI applications in finance. This paper provides a comprehensive review of AI-driven innovations in financial markets, banking services, and regulatory compliance while discussing ongoing challenges and future research directions.
Cryptocurrency represents a form of asset that has arisen from the progress of financial technology, presenting significant prospects for scholarly investigations. The ability to anticipate cryptocurrency prices with extreme accuracy is very desirable to researchers and investors. However, time-series data presents significant challenges due to the nonlinear nature of the cryptocurrency market, complicating precise price predictions. Several studies have explored cryptocurrency price prediction using various deep learning (DL) algorithms. Three leading cryptocurrencies, determined by market capitalization, Ethereum (ETH), Bitcoin (BTC), and Litecoin (LTC), are examined for exchange rate predictions in this study. Two categories of recurrent neural networks (RNNs), specifically long short-term memory (LSTM) and gated recurrent unit (GRU), are employed. Four performance metrics are selected to evaluate the prediction accuracy namely mean squared error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and root mean squared error (RMSE) for three cryptocurrencies which demonstrates that GRU model outperforms LSTM. The GRU model was implemented as a two-layer deep learning network, optimized using the Adam optimizer with a dropout rate of 0.2 to prevent overfitting. The model was trained using normalized historical price data sourced from CryptoDataDownload, with an 80:20 train-test split. In this work, GRU qualifies as the best algorithm for developing a cryptocurrency price prediction model. MAPE values for BTC, LTC and ETH are 0.03540, 0.08703 and 0.04415, respectively, which indicate that GRU offers the most accurate forecasts as compared to LSTM. These prediction models are valuable for traders and investors, offering accurate cryptocurrency price predictions. Future studies should also consider additional variables, such as social media trends and trade volumes that may impact cryptocurrency pricing.
Ibrahim Garba Kabo, Georgina N. Obunadike, Nuruddeen A. Samaila
Bitcoin, the leading cryptocurrency, has gained significant attention due to its high volatility and potential economic impact. Traditional financial forecasting models struggle to accurately predict Bitcoin prices due to its sensitivity to various factors, including market sentiment and macroeconomic conditions. Existing models primarily rely on historical price data, often neglecting external influences such as public sentiment and economic indicators like Gross Domestic Product (GDP). To address these limitations, this study explores a hybrid approach that integrates Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) models with sentiment analysis and GDP data to enhance Bitcoin price prediction accuracy. The study evaluates the predictive capabilities of these models under different scenarios. When trained on Bitcoin price data combined with sentiment analysis and GDP data, the ARIMA model achieved a Mean Absolute Error (MAE) of 2081.66, Root Mean Square Error (RMSE) of 2518.35, and an R-squared value of 0.9143. In comparison, when trained on Bitcoin data alone, it exhibited lower accuracy. The LSTM model demonstrated superior performance, achieving an MAE of 1253.24, RMSE of 1717.65, and an R-squared value of 0.9602 when incorporating sentiment and GDP data, significantly outperforming its standalone counterpart. The results highlight the effectiveness of integrating sentiment analysis and GDP data in cryptocurrency price prediction, demonstrating that hybrid models provide greater forecasting accuracy than traditional approaches. This study offers a robust framework for financial time series forecasting, aiding investors, analysts, and policymakers in making more informed decisions in the cryptocurrency market.
Cryptocurrencies have received a lot of attention from central banks, investors, and governments worldwide. The insufficiency of any method of political guideline and their market is far from "effective", so they want novel regulation methods shortly. From an econometric perspective, the technique underlying the growth of the cryptocurrencies' volatility was observed to demonstrate similarities and differences with other economic time series, e.g., foreign exchange yields. Accurate prediction of cryptocurrency price fluctuations is significant for effectual portfolio management and improves economic models by identifying potential risks and attacks. With the growing use of AI in various fields, its application in financial markets, especially cryptocurrencies and stocks, is an emerging research area. This study presents an Empirical Evaluation of Fuzzy Bidirectional Long Short-Term Memory with a Soft Computing-based Decision-Making Model for Predicting Volatility of Cryptocurrencies (FBLSTMSC-DMPVC) technique. The primary focus of the FBLSTMSC-DMPVC technique is to present a robust and intelligent framework for an advanced decision-making model to predict cryptocurrency volatility. Initially, the presented FBLSTMSC-DMPVC method performs the data preprocessing process using Z-score normalization to ensure all features are standardized and scaled. Furthermore, the fuzzy bidirectional long short-term memory (FBLSTM) method predicts cryptocurrency volatility. To enhance the hyperparameters of the FBLSTM technique, the improved carnivorous plant algorithm (ICPA) is employed. A wide range of simulation is accomplished to ensure the impact of the FBLSTMSC-DMPVC technique. The FBLSTMSC-DMPVC technique portrayed a superior MAPE value of 0.7939 for BTC, 0.8633 for ETH, 0.6187 for LTC, and 0.6667 for XRP, demonstrating its performance across various cryptocurrencies.
Albi Isufaj, Caio De Castro Martins, Marc Cavazza, Helmut Prendinger
This paper explores the applicability of Convergent Cross Mapping (CCM) and its extension, Time Delay Convergent Cross Mapping (TDCCM), to assess the causal relationships between Bitcoin, the S&P 500 index, and gold. Unlike conventional causality analysis methods, such as Granger causality or transfer entropy, CCM accounts for non-separable, weakly connected dynamic systems, and TDCCM explicitly incorporates time lags during cross-mapping, enabling the detection of complex causal relationships in systems with shared nonlinear behavior. This makes it particularly suitable for financial time series that often exhibit chaotic and nonlinear dynamics, particularly during periods of market instability. We integrate TDCCM with simplex projection and sequential locally weighted global linear map (S-map) algorithms, applying a sliding window approach to identify short time intervals characterized by high levels of nonlinearity and chaoticity. Using this approach, we uncovered a strong causal relationship between Bitcoin and the S&P 500 index during the onset of the COVID-19 pandemic. Our analysis reveals a bidirectional causal relationship between Bitcoin and the S&P 500 index, highlighting their interconnectedness during periods of heightened economic uncertainty. Furthermore, we find a unidirectional causal influence of Bitcoin on gold, reflecting Bitcoinâs evolving role as a macroeconomic indicator and its growing relevance as an alternative store of value. These findings provide insight into the dynamics between cryptocurrencies and traditional financial markets, particularly during periods of global economic disruption. ⢠We use Time Delay Convergent Cross Mapping (TD-CCM) to identify and quantify lagged causal interactions between financial time series (Bitcoin, Gold, and the S&P 500 index), which is a more recent and less explored method for Time Series causality. ⢠We report a comprehensive and replicable methodology to apply TD-CCM to non-linear TS, based on combining the S-Map and Prediction Decay algorithm with a sliding window technique, while validating with surrogate analysis. ⢠We show evidence of strong causal influence from Bitcoin to Gold and bidirectional causality between Bitcoin and the S&P 500 Index during significant economic events like the COVID-19 pandemic.
Social media has attracted society for decades due to its reciprocal and real-life nature. It influenced almost all societal entities, including governments, academics, industries, health, and finance. The Social Network generates unstructured information about brands, political issues, cryptocurrencies, and global pandemics. The major challenge is translating this information into reliable consumer opinion as it contains jargon, abbreviations, and reference links with previous content. Several ensemble models have been introduced to mine the enormous noisy range on social platforms. Still, these need more predictability and are the less-generalized models for social sentiment analysis. Hence, an optimized stacked-Long Short-Term Memory (LSTM)-based sentiment analysis model is proposed for cryptocurrency price prediction. The model can find the relationships of latent contextual semantic and co-occurrence statistical features between phrases in a sentence. Additionally, the proposed model comprises multiple LSTM layers, and each layer is optimized with Particle Swarm Optimization (PSO) technique to learn based on the best hyperparameters. The model's efficiency is measured in terms of confusion matrix, weighted f1-Score, weighted Precision, weighted Recall, training accuracy, and testing accuracy. Moreover, comparative results reveal that an optimized stacked LSTM outperformed. The objective of the proposed model is to introduce a benchmark sentiment analysis model for predicting cryptocurrency prices, which will be helpful for other societal sentiment predictions. A pretty significant thing for this presented model is that it can process multilingual and cross-platform social media data. This could be achieved by combining LSTMs with multilingual embeddings, fine-tuning, and effective preprocessing for providing accurate and robust sentiment analysis across diverse languages, platforms, and communication styles.
This paper focuses on automating the analysis of financial news for stocks and cryptocurrencies, thereby providing traders and analysts with actionable insights. In today's fast-paced markets, being up-to-date is the need of the hour; however, manual analysis takes time. To overcome this challenge, the paper employs Python and deep learning tools to make the process of collecting, summarizing, and conducting sentiment analysis on relevant news articles much easier. The key assets analyzed are popular stocks and cryptocurrencies, such as Tesla, Gamestop, Bitcoin, and Ethereum. The system uses Hugging Face's Pegasus model, which is a state-of-the-art transformer, to summarize lengthy financial news into concise, manageable summaries. This reduces the effort required to sift through large amounts of information while preserving essential details. Moreover, the system applies pre-trained sentiment analysis models to gauge the market's overall sentiment-positive, negative, or neutral-toward specific assets, thus making quick and informed trading decisions possible. The paper workflow includes automatically scraping web sources such as Google News and Yahoo Finance, cleaning and processing the data, and exporting results in structured CSV files for further analysis. The files include the ticker symbols, sentiment scores, confidence levels, URLs, and summary text. It is scalable and flexible so that users can input their stock tickers and run real-time analysis with changes in market conditions. Overall, this paper provides an efficient, end-to-end solution for financial news analysis, allowing users to make informed decisions with reduced time and effort on data gathering and interpretation.
The marketâs intrinsic volatility and the influence of outside variables like investor emotion and economic indicators make it difficult to predict cryptocurrency prices. This study investigates several data mining techniques to raise the predicting accuracy of bitcoin prices. We use sentiment analysis, machine learning methods, and time series analysis in combination to model price movements more effectively. Traditional forecasting methods, including ARIMA and GARCH, are used alongside advanced neural networks and ensemble methods to capture complex, non-linear patterns in the data. Additionally, the study highlights the critical role of feature engineering and the application of clustering strategies to enhance predictive model performance.The integration of these approaches demonstrates a marked improvement in forecasting outcomes, providing not only more accurate price predictions but also valuable insights into market dynamics. These findings can assist investors in developing more robust investment strategies, allowing them to better navigate the cryptocurrency marketâs risks. Our findings highlight how hybrid models can improve predictive capabilities and understanding market behavior in the evolving landscape of digital assets.
The cryptocurrency market is currently one of the most interesting areas for investment, attracting both experienced and casual investors. Although it can offer high returns, it also poses significant risks due to its high volatility. In this context, artificial intelligence, particularly through deep learning and machine learning algorithms, has played a key role in developing applications that provide investment advice, with the aim of maximizing returns and reducing investment risks. This study proposes a system for forecasting the closing prices of ten of the leading cryptocurrencies currently available in the market, presented in a web application capable of making predictions ranging from one to four hours. To achieve this, different models using various machine learning and deep learning algorithms were analyzed and tested, including Recurrent Neural Networks, time series analysis algorithms such as ARIMA, and even some more conventional regression algorithms. For algorithm comparison, minute step Bitcoin price data over a 30-day period was used to forecast prices 60 minutes ahead. Through extensive experimentation, the GRU neural network demonstrated superior predictive accuracy, achieving MAPE = 0.09\%, MSE = 5954.89, RMSE = 77.17, and MAE = 60.20. A web application was also developed, which integrates the best-performing model to provide real-time price predictions for multiple cryptocurrencies.
Cryptocurrencies do not have proper economic fundamentals. Consequently, economic variables cannot predict crypto prices. According to economic theory, cryptocurrencies are unbacked assets that are inherently unforecastable. However, a growing strand of literature suggests global crypto markets to be informationally inefficient. It implies the possibility of return predictability based on past information. Forecasting the allegedly unforecastable becomes feasible. Keeping it sophisticatedly simple, past infomation can be captured by autoregressive integrated moving average (ARIMA) processes of principal components. However, Principal Component Analysis (PCA) for crypto price series is due to their non-Gaussian property not applicable and requires the assumption of a stochastic trend model. Making use of the Central Limit Theorem, Independent Component Analysis (ICA) overcomes this deficiency. We show that ICA combined with ARIMA modeling more than triples the predictability of global crypto price dynamics. ⢠Crypto markets are found to be inefficient in the sense of majority games. ⢠ICA based ARIMA more than triples predictability of crypto price dynamics. ⢠ICA based ARIMA is most reliable for directional out-of-sample predictions.
ABSTRACT This study examines the connection between Bitcoin and global factors, including the VIX, the oil price, the US dollar index, the gold price, and interest rates estimated using the Federal funds rate and treasury securities rate, for forecasting analysis. Deep learning methodologies, including LSTM, GRU, CNN, and TFT, with machine learning algorithms such as XGBoost, LightGBM, and SVR, were employed to identify the optimal prediction model for the Bitcoin price. The findings indicate that the TFT model is the most successful predictive approach, with the gold price identified as the most relevant component in determining the Bitcoin price. After the gold indicator, the US dollar index was a substantial factor in the explanation of the Bitcoin price. The TFT model also included regulatory decisions and global events. It was estimated that the Bitcoin price was significantly influenced by the COVIDâ19 pandemic. After that, global climate events and China mining ban strongly affected the Bitcoin price. These findings indicate that regulatory decisions and global events determine the Bitcoin price in addition to macroeconomic factors. The VAR analysis was employed as a robustness check. The results indicate that gold and oil prices have a strong negative influence on Bitcoin, particularly in the long term. The paper has significant policy implications for investors, portfolio managers, and scholars.
Cryptocurrency is the most innovative financial and technological breakthrough of this generation. Investment in cryptocurrency grew from USD 11.18 billion in December 2016 to USD 2.147 trillion in April 2024; however, the rationality of investor exuberance is uncertain. This paper explores stakeholders' perceptions of cryptocurrency using a machine-learning approach based on artificial intelligence (AI). In particular, we employ a lexicon-based emotion-detection sentiment analysis to investigate stakeholder perceptions, using 2.3 million open-source data points. We divide the findings into positive, neutral, and negative stakeholder perception pillars based on factors such as trustworthiness in cryptocurrency, motives, cryptocurrency awareness and knowledge, ownership, socioeconomic characteristics of users, and usage. Our analysis reveals that 51 percent of the stakeholders have a positive perception of cryptocurrency, whereas 40 percent have a neutral perception and 9 percent a negative perception. After identifying the perceptions, we investigate the relationship between cryptocurrency prices and stakeholder perceptions using the autoregressive distributed lag (ARDL) framework with time-series data from August 2017 to July 2023. The long- and short-term results confirm that positive and negative perceptions have statistically significant effects on cryptocurrency prices. Individual investors comprise the largest share of those with a positive perception, as 54 percent have a positive view of cryptocurrency. Institutional investors, however, have the largest share of those with a neutral perception because of the lack of a well-established regulatory framework for cryptocurrency. However, 39 percent of institutional investors hold a positive perception is growing, a sign of a growing trend, as they are among the major investor groups with an interest in investing in crypto. Other stakeholders, such as the government, academia, and other miscellaneous groups, have a negative perception. Our results demonstrate that cryptocurrency has affected social change, social inclusion, and sustainability. Moreover, our findings offer social insights about crypto stakeholdersâ perceptions about the design of strategies to promote cryptocurrency and the establishment of a sustainable crypto ecosystem. ⢠The study investigates the stakeholders' perception towards cryptocurrency. ⢠AI-based sentiment analysis identifies that 51% of stakeholders have a positive perception towards cryptocurrency. ⢠The Autoregressive Distributed Lag model integrated with the UECM model was used to investigate the cointegration between cryptocurrency and stakeholder perception. ⢠A positive perception of cryptocurrency has a significant positive effect on its price.