Disaggregating renewable energy-growth nexus: W-ARDL and W-Toda-Yamamoto approaches

The research on the effects of economic growth on renewable energy faces two significant limitations that may impede the formulation of accurate and effective renewable energy policy. The first limitation pertains to the use of an aggregate methodology, which relies solely on aggregated data for analyzing the relationship between renewable energy and economic growth. The second limitation is the absence of a systematic search for empirical models that could minimize estimation errors. In light of these constraints, this study proposes a disaggregate analysis to examine the concurrent effects of the components of economic growth on the components of aggregate renewable energy. Second, the research introduces two methodologies: wavelet-based and error-adjusted variants of the Auto-Regressive Distributed Lag Model (ARDL) and the Toda-Yamamoto Causality Test. The study presents two major findings. First, it reveals substantial differences between the effects of aggregate growth and growth components on aggregate renewable energy and its components in the ten countries with the highest carbon emissions. Second, the study demonstrates that the predictive accuracy of wavelet-based and error-adjusted ARDL and Toda-Yamamoto tests (W-ARDL and W-Toda-Yamamoto) for all estimated models is significantly higher than that of non-wavelet-based tests. The study suggests prioritizing a disaggregate approach to the renewable energy-growth nexus and utilizing wavelet-based and error-adjusted econometric techniques to improve the accuracy and efficiency of renewable energy policy.