Analyzing Historical Trends and Predicting Market Sentiment in Digital Currency Using Time Series Decomposition and ARIMA Models on Crypto Fear and Greed Index Data

This study analyzes historical trends and predicts market sentiment in digital currencies using time series decomposition and ARIMA models, focusing on the Crypto Fear and Greed Index. The volatile nature of cryptocurrency markets, driven largely by investor sentiment, necessitates a thorough understanding of market mood to anticipate price movements and market dynamics. The research utilized time series decomposition to uncover significant trends and seasonal patterns within the sentiment data. The ARIMA model was applied to predict future sentiment, achieving a Mean Absolute Error (MAE) of 11.15 and a Root Mean Square Error (RMSE) of 13.30, indicating strong alignment with actual market behavior. Additionally, the study employed the Prophet model, which, although less precise with an MAE of 22.56 and RMSE of 24.98, provided valuable insights into the seasonal components of market sentiment. These results underscore the importance of sentiment analysis in digital currency markets, offering actionable insights for traders and investors. Limitations of the models are acknowledged, with suggestions for future research including the integration of additional data sources and more sophisticated modeling techniques to further refine sentiment predictions. This research contributes to the expanding body of knowledge on the role of sentiment analysis in financial markets, particularly within the dynamic field of digital currencies.