Analysis of predictions considering mineral prices, residential energy, and environmental risk: Evidence from the USA in COP 26 perspective

Abstract

The COP 26 member countries reaffirmed their commitment to the energy transition at the most recent 26th Conference of the Parties (COP 26) to reduce traditional energy and critical mineral as essential components to the growth of renewable energy transition. To fill the gap in the current literature, this study examined the effect of mineral price along with other essential factors like environmental risk, economic policy uncertainty, and residential energy use on energy transition. This study employs combined cubic support vector machine learning (CSVML) and Artificial Neural Networks (ANN) to predict energy transition through minerals prices, energy residential, economic policy uncertainty, and environmental risk. In doing so, the authors use the quarterly data from 1991Q1 to 2020Q4. The Cubic SVM has outperformed other models in terms of accuracy in predicting energy transition compared to ANN and other methods. In addition to comparing the findings of CSVM and ANN, it also gave policy implications. It has been predicted that minerals prices, low environmental risk, strong economic policy, and renewable energy use for residential purposes are the factor for a better energy transition in the US economy. As a result, our policy recommendations include maximizing mineral resources for clean energy transitions to achieve the COP 26 goal of a decarbonized or net-zero emissions trajectory for the twenty-first century.