Systemic tail dependence in disruptive technology ETFs & crypto assets: A partial correlation network

This study investigates the systemic dependence structure among Disruptive Technology Exchange-Traded Funds (ETFs) and Crypto Assets, focusing on dynamic, directional, and tail-specific connectedness. We explore static and dynamic dependence under normal and extreme return regimes from January 2020 to April 2025, utilizing the novel partial correlation-based network framework combined with quantile-specific connectedness measures. Unlike traditional GFEVD-based models, our methodology isolates direct, tail-sensitive interdependencies while filtering out standard shocks. Our results indicate strong state-dependent asymmetries, with significantly intensified co-movements in the lower tail of the return distribution. Dynamic connectedness spikes notably during key events such as the GME/Crypto retail rally, the Terra-Luna crash, and the NVIDIA-driven AI surge. The partial correlation framework further distinguishes itself by capturing sharper structural breaks and more meaningful directional asymmetries than traditional variance-based networks. Portfolio optimization results demonstrate that strategies minimizing connectedness in tail states outperform traditional variance- and correlation-based approaches. Our findings offer insights for tail-sensitive portfolio construction, ETF regulation, and crypto-market risk monitoring.