Towards Precision Economics: Unveiling GDP Patterns Using Integrated Deep Learning Techniques

Economic activities and gross domestic product (GDP) interplay is crucial for fostering sustainable growth and development. Therefore, this study provides enhancements, including the advancement of machine learning, such as the group method of data handling (GMDH) technique to predict gross domestic products in Tanzania. The GMDH model surpassed the particle swarm optimization-artificial neural network (PSO-ANN) and Catboost in both the training and testing phases, with correlation coefficient (R2), the lowest mean absolute error and root mean square error (RMSE).In GMDH training data has R2 = 0.9968, MAE = 0.2478, RMSE = 0.4978, testing dataset with R2 = 0.851, MAE = 0.2267, RMSE = 0.4967. GMDH reduces artificiality by rapidly learning training data and eliminating unnecessary neurons and prediction errors. Hence, Economic activities significantly impact GDP from merchandise trade by about 13.50%. At the same time, military expenditure and industry have a substantial influence of 6.37% and 2.92%, respectively. Exports of goods and services (1.08%) have the slightest effect on the gross domestic product estimation model. Our model has a tremendous potential impact on the adequacy of macroeconomic policy, providing tools that help to achieve macroeconomic and monetary stability at the global level and creating new methodological opportunities for GDP growth forecasting. The study suggests that researchers develop a deep learning model to detect and quantify company operations accurately using satellite imagery.