Title: Economic forecasting in a shifting world
Authors: David Hendry - University of Oxford (United Kingdom) [presenting]
Jennifer L Castle - Oxford University (United Kingdom)
Jurgen Doornik - University of Oxford (United Kingdom)
Abstract: Economic time series are subject to non-stationarities from both evolving stochastic trends and sudden distributional shifts, often unanticipated such as pandemics. Modelling and forecasting difficulties are exacerbated by the latency in data provision, usually followed by substantive revisions and occasional changes to data measurement systems. Despite a large body of theory, economists have imperfect and incomplete knowledge of their data generating processes from changing human behaviour, so must search for reasonable empirical modelling approximations. Despite such problems, forecasts of likely future outcomes and their uncertainties are essential to plan and adapt as events unfold, over varying horizons for different decisions. We consider how these features shape the formulation and selection of econometric models for forecasting, and apply our tools to forecasting top-income shares.