Dirty versus renewable energy consumption in China: a comparative analysis between conventional and non-conventional approaches

Abstract

This study uses two empirical approaches to explore the asymmetric effects of oil and coal prices on renewable energy consumption (REC) in China from 1970 to 2019. As a conventional approach, we used the nonlinear autoregressive distributed lags (NARDL) model, while machine learning was used as a non-conventional approach. The empirical findings of the NARDL indicate that oil and coal price fluctuations have a significant effect on REC for both the short and long term. The results of the non-conventional approaches based on machine learning indicated that the SVM model was more efficient than the KNN model in terms of accuracy, performance, and convergence. Referring to the SVM model findings, the results show that an increase in the coal price has a higher ability to predict REC than the oil price. As a robustness check, we also find that an increase in Brent prices significantly decreases REC. The findings of this study support the view that there is a substitution effect from oil to coal before initiating the use of renewable energy in China.