In a world where breaks are unanticipated yet occur intermittently, it is easy to understand why economic forecasting has a poor track record of missing major changes, and being systematically wrong for prolonged periods. We are developing improved forecasting methods; approaches to try and forecast breaks, or failing that, forecasting during breaks; and building robust forecasting devices. Accurate and timely knowledge about the current state of the economy is important for expectations formation, accurate forecasting and policy making, but official data are released with delays and often revised. Nowcasting aims to capture a better prediction of the current state of the economy by using large datasets on variables released more frequently and finding statistical links between these leading indicators and macroeconomic numbers such as growth and inflation.
People: David Hendry, Jennifer Castle, Michael P. Clements and Andrew Martinez