11th European Central Bank Conference on Forecasting Techniques
Macroeconomic forecasting in abnormal times
Tuesday, 15 and Wednesday, 16 June 2021
European Central Bank
Online event
The biennial ECB Conference on Forecasting Techniques provides a forum for new theoretical and applied work on forecasting. The conference brings together experts from different fields to exchange new ideas on macroeconomic and financial forecasting. It looks for new ways to put insights from econometric and statistical theory into practice, with a particular focus on the current macroeconomic environment.
This year’s conference is dedicated to the subject of forecasting in abnormal times. Presentations look e.g. at robust forecasting in the presence of non-linearities, structural breaks and tail events (such as COVID-19).
In the context of the conference, the ECB organised a paper competition for PhD students with a research interest in forecasting. The winning entry – Varlam Kutateladze’s paper entitled “The Kernel Trick for Nonlinear Factor Modeling” – will open the tasting session on the first day of the conference.
Conference summary
The coronavirus (Covid-19) pandemic shock, new or shifting economic trends and revisions to major central banks’ monetary policy make macroeconomic forecasting a challenging task. Researchers are currently advancing primarily on two fronts, either by sheltering linear models from extreme events or by explicitly modelling the dynamics of such events. New approaches and methods are developing rapidly, partly inspired by big data and machine learning techniques.
Read more about the main takeaways from this conference in the VoxEU column “After floods and pandemics: How to obtain a meaningful forecast".
Programme
Times are Central European Time (UTC+2)
* indicates the presenter
- 12:00
- Registration
-
Chair: Siem Jan Koopman, Vrije Universiteit Amsterdam
- 12:55
-
Welcome
- 13:00
-
Welcome address
Philip R. Lane, European Central Bank
- 13:15
-
Paper 1: Expecting the unexpected: economic growth under stress
- Esther Ruiz*, Universidad Carlos III de Madrid
- Gloria Gonzalez-Rivera, UC Riverside
- Vladimir Rodríguez-Caballero, ITAM
Discussant: Simone Manganelli, European Central Bank
- 14:00
-
Paper 2: Tail forecasting with multivariate Bayesian additive regression trees
- Florian Huber*, Universität Salzburg
- Todd E. Clark, Federal Reserve Bank of Cleveland
- Gary Koop, University of Strathclyde
- Massimiliano Marcellino, Bocconi University
- Michael Pfarrhofer, Universität Salzburg
Discussant: Herman van Dijk, Erasmus University Rotterdam
- 14:45
- Break
- 15:00
First paper tasting
- 15:02
-
Taste 1: The kernel trick for nonlinear factor modeling
Varlam Kutateladze*, UC Riverside, Winner of the paper competition for PhD students
- 15:09
-
Taste 2: Variable selection and forecasting in high dimensional linear regressions with parameter instability
- Mahrad Sharifvaghefi*, University of Pittsburgh
- Alexander Chudik, Federal Reserve Bank of Dallas
- Hashem Pesaran, University of Southern California
- 15:16
-
Taste 3: The macroeconomy as a random forest
Philippe Goulet Coulombe*, University of Pennsylvania
- 15:23
-
Taste 4: An interpretable machine learning workflow with an application to economic forecasting
- Andreas Joseph*, Bank of England
- Marcus Buckmann, Bank of England
- Helena Robertson, Financial Conduct Authority
- 15:30
-
Taste 5: Understanding growth-at-risk: a Markov-switching approach
- Francesca Loria*, Federal Reserve Board
- Dario Caldara, Federal Reserve Board
- Pablo Cuba-Borda, Federal Reserve Board
- Danilo Cascaldi-Garcia, Federal Reserve Board
- 15:37
-
Taste 6: High-frequency monitoring of growth-at-risk
- Laurent Ferrara*, SKEMA Business School
- Matteo Mogliani, Banque de France
- Jean-Guillaume Sahuc, Banque de France
- 15:44
-
Q&A based on chat questions
- 16:00
- Break
- 16:15
-
Keynote speech: Large Bayesian VARs for forecasting: shrinkage priors, stochastic volatility and computation
Joshua Chan, Purdue University
- 17:15
- Break
- 17:30
-
Paper 3: Addressing COVID-19 outliers in BVARs with stochastic volatility
- Elmar Mertens*, Deutsche Bundesbank
- Andrea Carriero, Queen Mary University of London
- Todd E. Clark, Federal Reserve Bank of Cleveland
- Massimiliano Marcellino, Bocconi University
Discussant: Michele Lenza, European Central Bank
- 18:15
-
Paper 4: The time-varying evolution of inflation risks
- Dimitris Korobilis*, University of Glasgow
- Alberto Musso, European Central Bank
- Bettina Landau, European Central Bank
- Anthoulla Phella, European Central Bank
Discussant: Ana Galvão, Warwick Business School
- 19:00
-
End of day 1
- 12:00
- Registration
- Chair: Todd E. Clark, Federal Reserve Bank of Cleveland
- 12:55
- Welcome
- 13:00
-
Paper 5: Modelling volatility cycles: The (MF)2 GARCH model
- Christian Conrad*, Universität Heidelberg
- Robert F. Engle, New York University
Discussant: Christian Brownlees, Universitat Pompeu Fabra
- 13:45
- Break
- 13:50
Second paper tasting
- 13:52
-
Taste 7: Advances in nowcasting economic activity: secular trends, large shocks and new data
- Ivan Petrella*, Warwick Business School
- Juan Antolín-Díaz, London Business School
- Thomas Drechsel, University of Maryland
- 13:59
-
Taste 8: Capturing GDP nowcast uncertainty in real time
Paul Labonne*, King's College London
- 14:06
-
Taste 9: Nowcasting 'true’ monthly US GDP during the pandemic
- Aubrey Poon*, University of Strathclyde - presentation slides
- Gary Koop, University of Strathclyde
- Stuart McIntyre, University of Strathclyde
- James Mitchell, Federal Reserve Bank of Cleveland
- 14:13
-
Taste 10: The value of robust statistical forecasts in the COVID-19 pandemic
- Jennifer Castle*, University of Oxford
- Jurgen Doornik, University of Oxford
- David Hendry, University of Oxford
- 14:20
-
Taste 11: Forecast comparison tests under fat-tails
- Mamiko Yamashita*, Toulouse School of Economics
- Jihyun Kim, Toulouse School of Economics
- Nour Meddahi, Toulouse School of Economics
- 14:27
-
Taste 12: Combining Bayesian VARs with survey density forecasts: does it pay off?
- Francesco Ravazzolo*, Free University of Bozen-Bolzano
- Marta Bańbura, European Central Bank
- Federica Brenna, KU Leuven
- Joan Paredes, European Central Bank
- 14:34
-
Q&A based on chat questions
- 14:45
- Break
- 15:00
-
Keynote speech: Forecasting with Bayesian SVARs
Christopher Sims, Princeton University
- 16:00
-
Paper 6: Robust forecasting
- Frank Schorfheide*, University of Pennsylvania
- Timothy Christensen, New York University
- Hyungsik Roger Moon, University of Southern California
Discussant: Tatevik Sekhposyan, Texas A&M University
- 16:45
- Break
- 17:00
-
Panel discussion on macroeconomic forecasting:
Our present challenges and the way ahead- Joshua Chan, Purdue University
- Christopher Sims, Princeton University
- Matt Taddy, Amazon
Moderator: Lucrezia Reichlin, London Business School
- 18:00
Closing remarks
- 18:05
-
End of conference
Audiovisual notice: Images and video recordings may be published online.
Please note that this programme may be subject to change without notice.
General information
English
- Elena Bobeica, European Central Bank
- Gabriel Perez Quiros, European Central Bank
- Gerhard Rünstler, European Central Bank
- Georg Strasser, European Central Bank
Carol Sue Lehmann
Directorate General Research
Monetary Policy Research Division