Analyzing Price Volatility of Hedera Hashgraph Using GARCH Models: A Data Mining Approach

Abstract

This study employs the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) model to analyze the volatility dynamics of Hedera Hashgraph, a prominent cryptocurrency. Using a dataset of 1,901 daily price observations, we investigate the presence of volatility clustering and the persistence of market shocks, which are hallmarks of financial markets. The GARCH(1,1) model demonstrates robust performance, with a Log-Likelihood of 2927.50, AIC of -5846.99, and BIC of -5824.79, confirming its suitability for volatility estimation. Key findings reveal significant volatility clustering, with alpha (α = 0.20) and beta (β = 0.78) indicating moderate sensitivity to recent shocks and high persistence of volatility, respectively. Visualizations of conditional volatility and historical price data highlight the inverse relationship between price stability and volatility, with high volatility periods accounting for 33% of the dataset. These insights underscore the importance of real-time volatility monitoring for risk management and investment strategies. The study concludes by suggesting future research directions, including the integration of GARCH models with machine learning techniques and the exploration of external factors influencing cryptocurrency price dynamics.