Prediction of CO2 emissions in Saudi Arabia using genetic algorithms based on Grey Model GM (1,1)

Abstract

Examining the economic elements of gas emissions and their effects is critical, especially given the present upward trend in its emission volume. Hence, this study aimed to predict CO2 emissions using Genetic Algorithms (GAs) based on the Grey Model GM (1,1) in Saudi Arabia. This involves constructing a high precision predicting model to find the optimal solution for the Grey Model using GAs (optimization method) based on the prediction error minimization. This study uses CO2 emissions in Saudi Arabia from 2007 to 2016 (in - sample and out - sample data sets) to evaluate the proposed prediction model, and which demonstrates how GAs may be used to improve the Grey theory. The mean absolute percentages error is criterion with which to compare the various forecasting models results. The results show that genetic algorithm based on Grey model GAGM (1,1) has highly accurate prediction than Grey model GM (1,1), which means that GAGM (1,1) can improve the precision of Grey forecasting model significantly. The suggested model is a highly accurate to predict CO2 emissions for Saudi Arabia in modelling and forecasting. So, the study results may be useful to the government in the development of the future economic policies.