Scholarlink

Solid biofuels are a crucial energy source for Sub-Saharan Africa, with a significant portion of the population relying on them for cooking and heating. The prediction of Gross Domestic Product (GDP) is a critical task for economic planning and policy making. Traditional statistical methods have been widely used for this purpose; however, the advent of machine learning techniques offers new opportunities for improving predictions. Several machine learning models are used to determine their effectiveness in predicting GDP per Capita. The models’ performance was assessed based on many metrics. Support Vector Regression achieved the lowest MAE of 0.81 and the highest R2 of 0.98. Ridge and Lasso Regression also performed well with an MAE of 0.89 and 0.88, respectively, and a high R2 of 0.98. In contrast, Decision Tree and Random Forest exhibited higher MAE values and lower R2, while XGBoost and TabNet provided intermediate performance metrics. Accurate GDP forecasts enable governments and organizations to make informed decisions.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning Techniques for GDP Per Capita Prediction in Developing Countries: A Case Study of Sub-Saharan Africa

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Hanane Zermane

摘要

Solid biofuels are a crucial energy source for Sub-Saharan Africa, with a significant portion of the population relying on them for cooking and heating. The prediction of Gross Domestic Product (GDP) is a critical task for economic planning and policy making. Traditional statistical methods have been widely used for this purpose; however, the advent of machine learning techniques offers new opportunities for improving predictions. Several machine learning models are used to determine their effectiveness in predicting GDP per Capita. The models’ performance was assessed based on many metrics. Support Vector Regression achieved the lowest MAE of 0.81 and the highest R2 of 0.98. Ridge and Lasso Regression also performed well with an MAE of 0.89 and 0.88, respectively, and a high R2 of 0.98. In contrast, Decision Tree and Random Forest exhibited higher MAE values and lower R2, while XGBoost and TabNet provided intermediate performance metrics. Accurate GDP forecasts enable governments and organizations to make informed decisions.