Energy indices are essential for analyzing global energy trends and their economic and societal impacts. These indices guide investment decisions, set performance standards, and help mitigate risks in energy assets. They are crucial for monitoring energy security, identifying supply chain vulnerabilities, and shaping policies to improve efficiency, sustainability, and fair energy distribution. This study employs mathematical-based artificial intelligence models to forecast the World Energy Index (WEI) using nine parameters, including Iraq’s GDP. Mathematical-based artificial intelligence (AI) models that rely on extensive data have demonstrated significant promise in utilizing enormous datasets and capturing intricate correlations for the goal of making predictions. Twelve AI models were tested, with the Gaussian Process Regression-Exponential Kernel model showing superior performance, with a training result MSE=0.011501, RMSE=0.107243, VAF= 0.9998, MAE=0.0638, MAPE=0.0800, R 2 =0.9999, PEI=1.9197, and testing result MSE=298.5360, RMSE=17.2782, VAF= 0.9608, MAE=6.1548, MAPE=3.0699, R 2 =0.9660, PEI=-1.1430. The findings offer valuable insights for policymakers and investors to make informed decisions in energy markets. Mathematical-based artificial intelligence (AI) models that rely on extensive data have demonstrated significant promise in utilizing enormous datasets and capturing intricate correlations for the goal of making predictions.
Abbas et al. (Sun,) studied this question.