Nowcasting GDP in Romania: A Dynamic Factor Model Approach

Abstract This paper explores the use of dynamic factor models for nowcasting gross domestic product in Romania, focusing on their potential to enhance short-term forecasting accuracy and support economic policy decisions. Nowcasting models have become essential in macroeconomic analysis due to their ability to integrate high-frequency data and address delays in official statistical releases. The study builds on the growing body of research demonstrating the effectiveness of factor models in capturing latent economic dynamics through large datasets. The methodology applies a dynamic factor model framework, grouping variables into four categories: Global, Soft, Real, and Labor factors. The model is estimated using the Expectation Maximization algorithm, and common factors are extracted through Kalman filtering. Monthly and quarterly indicators from January 2007 to December 2024, including trade, industrial production, and labor market data, are employed. The research questions center on the accuracy of short-term gross domestic product projections and the relative contribution of different factor categories. The findings reveal that the model effectively captures economic trends, with strong correlations between model estimates and actual data. This study underscores the significance of dynamic factor models in real-time economic assessment for emerging markets. By addressing publication delays and integrating diverse data sources, the proposed approach offers valuable insights for policy-making and macroeconomic analysis, contributing to the ongoing development of advanced forecasting techniques.