Trade in Value-Added (TiVA) indicators are essential tools for understanding how countries are integrated into global supply chains. However, these indicators are published with considerable lags, limiting their usefulness for policymakers who need to respond to rapidly changing economic conditions. This paper aims to nowcast four TiVA indicators across a panel of 41 economies, both at the aggregate (economy-wide) level and for 24 industry sectors. The analysis relies on gradient boosted trees and other machine-learning techniques in a panel-data setting. The resulting nowcasting algorithms improve forecasts compared to an autoregressive benchmark model and yield relatively low prediction errors at a one- and two-year horizons, although model performance varies across countries and sectors.
Mourougane et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: