Return Spillovers Among Conventional Energy, New Energy Subsectors, ESG Assets and Gold: A Quantile Time–Frequency Connectedness Network Approach

Geopolitical conflict and the energy transition can alter the dependence between conventional energy, new energy, ESG assets, and gold. This study examines ten Chinese-market exposures using quantile time–frequency connectedness and network analysis. Mean rolling total connectedness is 60.406% at the median, 86.868% in the lower tail, and 86.664% in the upper tail. Wind power and photovoltaics have the largest median outgoing strengths, but directional positions differ across sectors and frequency bands. Short-frequency contributions dominate the baseline estimates; the long-frequency share is larger in the lower tail. Unknown-break tests do not reject constant-mean stability at 5%, and fixed-date confidence intervals include zero; neither result identifies a permanent causal effect of the Russia–Ukraine conflict. A separate portfolio evaluation over 1500 dates shows that minimum-variance allocations reduce realized variance, although weights and performance depend strongly on the gold proxy. Coal/gold-only and positive-NET allocations have different risk–return tradeoffs, and a net-receiver position alone does not establish hedge effectiveness. The findings support monitoring dependence across return states and testing explicit allocation rules, subject to common exposures, preprocessing, and instrument verification.