Forecasting Iraqi GDP Using Artificial Intelligence

Forecasting economic indicators like Gross Domestic Product (GDP) is crucial for planning and decision-making by policymakers, investors, and businesses. Traditional econometric models, including time series and regression analyses, often fail to capture the complex, non-linear dynamics in economic data. This paper explores the application of artificial neural networks (ANNs), specifically a multilayer perceptron (MLP) model with three hidden layers, to forecast Iraq's GDP. The volatility of Iraq's economy, heavily influenced by oil revenues and geopolitical instability, presents unique challenges. Using quarterly GDP data from 2000 to 2020, the ANN model was better at predicting the future, with an R-squared value of 0.996 and a mean absolute percentage error (MAPE) of 3.97%. These results indicate high accuracy and reliability, underscoring the potential of ANNs to enhance economic forecasting in developing and resource-dependent economies. The findings offer critical insights for economic planning and policy formulation, particularly in settings similar to Iraq's. This study not only contributes to a deeper understanding of AI applications in economic analysis but also opens up avenues for further exploration of AI-based models in other volatile economic environments.