Cryptocurrency Price Forecasting Using ARIMAX: Conceptual Framework

Abstract

Cryptocurrency, a digital or virtual currency secured by cryptography, has become a dynamic and volatile asset class, presenting both opportunities and challenges for traders and investors. This study aims to develop a web-based application prototype for forecasting Bitcoin prices using the Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) model. The ARIMAX model, known for its ability to integrate external factors into time series forecasting, is applied to historical Bitcoin price data, utilizing the Crypto Fear and Greed Index as an exogenous variable. By addressing the inherent volatility and unpredictability of Bitcoin's price movements, this study seeks to enhance the reliability of price predictions. Model performance will be evaluated using metrics such as Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The resulting web application will deliver real-time price forecasts and analyses, empowering users to make informed decisions and manage risks more effectively in cryptocurrency trading. This research ultimately aims to contribute to advancements in predictive modeling techniques within financial technology, providing traders and investors with a valuable tool for navigating the complex cryptocurrency market.