This research provides an empirical analysis of the dynamic relationship between clean and dirty energy markets. Specifically, we use Brent crude, West-Texas-Intermediate (WTI) crude, OPEC oil, Crude oil Oman and Crude Oil Dubai to denote dirty energy markets and use the S&P Global Clean Energy Index and WilderHill New Energy Global Innovation Index as a representative of the clean energy market. The time-frequency wavelet's multiple cross-correlation and cross-quantilogram correlation are used as estimation techniques to examine time-dependent wavelet cross-correlation and directional predictability, respectively. We use daily returns spanning from November 2013 to September 2020. Findings from the cross-quantilogram correlation (CQC) results suggest heterogeneous quantile dependence dynamics from clean energy markets to dirty energy markets. Additionally, findings from the cross-quantile correlation results reveal positive and negative directional predictability between clean and dirty energy markets in high, medium and low quantile ranges. Second, results from the time-frequency wavelets multiple cross-correlation approach suggest that clean and dirty energy markets are marginally integrated at the lowest frequencies, with dirty energy emerging as a predictive power of clean energy. In addition, we also find that the co-movements between the clean and dirty energy sources are volatile in the medium and long term, thus reducing the medium- and long-term diversification sphere. These findings are relevant for portfolio managers and clean energy producers.

本研究对清洁能源市场和污染能源市场之间的动态关系进行了实证分析。具体来说，我们使用布伦特原油、西德克萨斯中质原油 (WTI) 原油、欧佩克原油、阿曼原油和迪拜原油来表示肮脏的能源市场，并以标准普尔全球清洁能源指数和 WilderHill 新能源全球创新指数为代表的清洁能源市场。时频小波的多重互相关和互量化图相关被用作估计技术，分别检验时间相关小波互相关和方向可预测性。我们使用从 2013 年 11 月到 2020 年 9 月的每日回报率。交叉数量图相关 (CQC) 结果表明，从清洁能源市场到污染能源市场存在异质分位数依赖动态。此外，交叉分位数相关结果的发现揭示了高、中、低分位数范围内清洁能源和脏能源市场之间的正向和负向可预测性。其次，时间-频率小波多重互相关方法的结果表明，清洁能源和脏能源市场在最低频率上略微整合，脏能源成为清洁能源的预测能力。此外，我们还发现，清洁能源与污染能源之间的联动在中长期内具有波动性，从而降低了中长期多元化领域。