Forecasting State Macroeconomic Indicators with Artificial Intelligence Tools

In an era of rapid advances in artificial intelligence (AI), innovation increasingly shapes many domains of human activity. Public administration is no exception: across the world, governments are adopting AI for forecasting and planning-from the Baltic states’ e-Governance platforms to Southeast Asia’s “smart state” initiatives (e.g., Singapore) and national AI programs in the Middle East. The shared goal is to raise the quality of forecasts and to make planning, crisis management, and economic policy more responsive to change.

This paper presents a practical, step-by-step methodology for producing targeted, one-year forecasts of state macroeconomic indicators. The approach combines time-series analysis and neural networks with deep learning (BiLSTM + Attention), using efficient, iterative procedures that systematically increase forecast accuracy.