Decomposing the persistence structure of Islamic and green crypto-currencies with nonlinear stepwise filtering

Abstract

• Islamic and green crypto-currency series are investigated for the first time in econophysics. • We introduce a novel multi-step nonlinear decomposition approach. • The (dis)similarities in chaotic measures of Islamic, green & common cryptos are probed. • Heterogeneous multi-fractal and long memory patterns are observed. • The short-term predictability is pronounced compared to conventional crypto-currencies. The multi-fractal chaotic dynamics of Islamic and Green crypto-currency series are investigated for the first time in econophysics literature. Specifically, we decompose and analyse the temporal signals of prices, returns, volume and volatility of Islamic and Green cryptos vis-à-vis conventional ones in a comparative manner. We introduce a multi-step resolution approach based on detrended fluctuation analysis, Generalized Hurst and Lyapunov exponents as well as fractionally integrated conditional heteroskedasticity. Moreover, various tests are employed to investigate the statistical significance of any (dis)similarities of long memory patterns, multi-fractality measures and chaotic dynamics observed among Islamic, Green and conventional crypto-currency markets. Our findings suggest that while the returns of Islamic and green crypto-currencies exhibit anti-persistent dynamics, their price, volatility and volume series embed high persistence compared to the conventional crypto-currencies. Further statistical testing indicates that the distributions of the chaotic parameter estimates are significantly different versus common crypto-currencies, a fact that reveals heterogeneity in multi-fractality and long memory patterns. As the Islamic and Green cryptos exhibit a distinct and more profound chaotic behaviour compared to conventional ones, their short-term predictability could further induce financial agents.