Journal article

Evaluating the use of realtime data in forecasting output levels and recessionary events in the USA

C Aristidou, K Lee, K Shields

Journal of the Royal Statistical Society Series A: Statistics in Society | Wiley | Published : 2019


The paper proposes a modelling framework and evaluation procedure to judge the usefulness of realtime data sets incorporating past data vintages and survey expectations in forecasting. The analysis is based on ‘metamodels’ obtained by using model averaging techniques and judged by various statistical and economic criteria, including a novel criterion based on a fair bet. Analysing US output data over 1968, quarter 4–2015, quarter 1, we find that both elements of the realtime data are useful with their contributions varying over time. Revisions data are particularly valuable for point and density forecasts of growth but survey expectations are important in forecasting rare recessionary events..

View full abstract

University of Melbourne Researchers


Awarded by Australian Research Council

Funding Acknowledgements

We received helpful comments from the referees and Associate Editor and from participants at the Computational Economics and Finance Conference, Canada, the Reserve Bank of New Zealand Nowcasting and Model Combination Workshop, a Philadelphia Fed Workshop on 'Real time data analysis', the Conference of the Centre for International Research on Economic Tendency Surveys, Denmark, and a Reserve Bank of Australia seminar. We are grateful for funding received from the Australian Research Council (grant DP0988112).