Predicting the Youth Unemployment Rate in Iraq Until 2035 Using Artificial Intelligence

Youth unemployment significantly impacts the socio-economic stability and growth of developing countries, particularly Iraq. This study aims to predict the youth unemployment rate in Iraq until 2035 using advanced AI techniques such as autoregressive integrated moving averages (ARIMA) and recurrent neural networks (RNNs). We developed and tested both ARIMA and RNN models using data from 2000 to 2023. The ARIMA (1, 1, 0) model showed strong predictive ability with an R-squared value of 0.954, an RMSE of 1.350, and a MAPE of 3.113. In contrast, the RNN model, enhanced by Long Short-Term Memory (LSTM) networks, exhibited superior performance with an RMSE of 0.113, a MAPE of 0.4888, and an R-value of 0.9998. Predictions reveal a rising trend in youth unemployment, expected to increase from 33.22% in 2024 to 45.63% by 2035. These findings underscore the effectiveness of AI-driven models in providing reliable forecasts, which are crucial for policymakers to devise targeted interventions. Additionally, accurate forecasting aligns with several United Nations Sustainable Development Goals, particularly Goal 8 (Decent Work and Economic Growth), and supports efforts in poverty reduction and educational alignment with labor market needs. Future research will integrate more socio-economic variables and explore further model enhancements to improve predictive accuracy. This study is valuable for its practical application in socio-economic planning and forecasting to mitigate youth unemployment in Iraq.