This thesis focuses on the challenge of short-term economic forecasting under conditions of very limited data availability due to the Russian invasion of Ukraine. This has led to a gap in the publication of statistical data, causing traditional forecasting models to lose their effectiveness, particularly in the area of regional forecasts. The study evaluates the forecasting accuracy of MF-FAVAR and SARIMAX. The results show that nowcasting models can be developed even with sparse data. This thesis contributes to the literature on economic monitoring during times of crisis and provides a model for real-time forecasting in post-conflict reconstruction.