Cryptocurrency Price Forecasting in a Volatile Landscape: SARIMAX Modeling and Short-Term Strategies

Cryptocurrencies, most notably Bitcoin, have experienced a significant increase in popularity, garnering the interest of both investors and scholars. The present study aims to investigate and forecast the prices of Bitcoin. The focus lies on the essential aspects of data preprocessing, exploratory data analysis, and forecasting methodologies. The dataset undergoes thorough cleaning procedures to ensure meticulousness, followed by an exhaustive exploration of the data through various analytical techniques. This study provides valuable insights into pricing trends, seasonality patterns, and relationships within the dataset. The research utilizes a range of models, such as ARIMA for short-term prediction, LSTM neural networks for intricate pattern detection, and hybrid models for enhanced resilience. The evaluation of model performance is conducted by using metrics such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE), while also employing out-ofsample testing to evaluate the model's ability to generalize. The results of this study provide a comprehensive analysis of the advantages and disadvantages associated with each technique, thereby offering significant insights for investors aiming to effectively navigate the cryptocurrency market. Furthermore, this highlights the possibility of applying this analysis to additional cryptocurrencies and improving forecasting models through the incorporation of sentiment analysis and macroeconomic indicators.