\Forecasting Carbon Dioxide Emissions for D-8 Countries by Artificial Neural Networks: Levenberg-Marquardt Algorithm

In recent years, growth, development, and sustainability-oriented lifestyles have created a separate problem, especially for developing countries. In this study, GDP per capita, energy consumption per capita (renewable, fossil, total), and urban population for D-8 countries (Indonesia, Bangladesh, Iran, Egypt, Malaysia, Pakistan, Nigeria, and Turkiye), all of which are developing countries, between 1990 and 2020. The carbon dioxide emission rates of the countries were estimated using the artificial neural networks (ANN) method by using the data of increase and carbon dioxide emission. In the ANN model established in the study, 70% of the database data was randomly divided into training, 15% validation and 15% test data. This artificial neural network is trained with the Levenberg-Marquardt method. Regression R values, one of the performance indicators of the model, were determined as 0,99 for training data, 0,97 for validation data and 0,99 for test data. The regression R-value for all data used in the model was determined as 0,99

Journal

SOSYOEKONOMI

Year

2024

Volume

32

Issue

59

Pages

363-382

ISSN

1305-5577

DOI

10.17233/sosyoekonomi.2024.01.16

WoS UID

WOS:001163472900003