Examining Long-Term and Short-Term Cryptocurrency Interactions: A Wavelet and Quantile Regression Approach

Abstract

Our research re-examines the dynamics of major cryptocurrencies in response to recent economic and geopolitical events, focusing specifically on the Covid-19 pandemic period. Employing wavelet analysis and quantile regression methods, we examine cryptocurrency behavior before, during, and after the pandemic, using the Least Asymmetric Daubechies (LA8) wavelet function to decompose log-returns into frequency scales. Additionally, wavelet coherence and quantile-on-quantile regression techniques analyze daily price data from July 2017 to April 2023. Results uncover a robust long-term relationship among cryptocurrencies with diminishing medium-term correlations. Bitcoin shows synchronization with major cryptocurrencies, excluding Tether, and BTC-ETH and BTC-LTC pairs exhibit interconnected dynamics alongside fundamental links. Empirical evidence highlights Bitcoin&quots heterogeneous relationship, being more responsive to positive extremes than negatives. Our study&quots timeframe limitation (July 2017 to April 2023) restricts insights into longer-term trends, and methodological choices like wavelet analysis may not fully capture cryptocurrency dynamics, suggesting potential for alternative interpretations. Nonetheless, findings suggest significant implications for investment strategies, emphasizing temporal dynamics within cryptocurrency markets.