A Machine Learning Approach to Predict Economic Freedom Index

Abstract

In this study, machine learning methods are used to estimate a nation's economic freedom index. Economic growth and development have been proven to be strongly correlated with the economic freedom index, which measures a nation's economic laws and regulations. This study used a dataset that included the Economic Freedom Index and several economic variables for 162 countries. The data underwent processes such as outlier removal, encoding categorical variables, and filling in missing values. The performance of a machine learning model was enhanced through hyperparameter tuning after it had been trained using a variety of techniques, including decision trees, random forests, and XGBoost. The outcomes demonstrate that the XGBoost algorithm outperformed other models. It had a 92% accuracy rate in predicting a nation's economic freedom index. In order to help policymakers improve a nation's economic laws and regulations, this study illustrates the potential of employing machine learning techniques to anticipate economic freedom.