Advances in nowcasting the growth rate of the ICT sector

The information and communication technology (ICT) sector is a key driver of modern economies, yet official measures of ICT value added are published with multi-year lags, limiting their usefulness for real-time monitoring. This paper extends the OECD's machine learning nowcasting model for ICT sector growth by refining its Google Trends feature set, updating it using the latest OECD Structural Analysis (STAN) Database, and introducing a selective pooling strategy based on countries' structural positions within global ICT value chains, derived from Trade in Value Added (TiVA) data. Results show that pooling countries with similar value-chain linkages improves prediction accuracy relative to global pooling. The findings suggest that the pooling strategy should be treated as a model-selection problem informed by economic structure, offering an empirically grounded framework for tracking ICT sector growth.