Predictive Density Combination Using a Tree-Based Synthesis Function

Seminar summary

Bayesian predictive synthesis (BPS) provides a method for combining multiple predictive distributions based on agent/expert opinion analysis theory and encompasses a range of existing density forecast pooling methods. The key ingredient in BPS is a “synthesis” function. This is typically specified parametrically as a dynamic linear regression. In this paper, we develop a nonparametric treatment of the synthesis function using regression trees. We show the advantages of our tree-based approach in two macroeconomic forecasting applications. The first uses density forecasts for GDP growth from the euro area’s Survey of Professional Forecasters. The second combines density forecasts of US inflation produced by many regression models involving different predictors. Both applications demonstrate the benefits – in terms of improved forecast accuracy and interpretability – of modeling the synthesis function nonparametrically.

Read the paper online

How to attend this seminar

This seminar will take place on Thursday 30 May 2024 at 3pm.

We welcome you to join us online via zoom.

This seminar is free to attend with no need to register in advance.

Speaker bio

Professor James Mitchell

James Mitchell is a vice president in the Research Department at the Federal Reserve Bank of Cleveland. He leads the macroeconomic forecasting group, which is responsible for developing models for economic forecasting and policy analysis. His research interests are in macroeconomic modeling and forecasting. Before joining the Cleveland Reserve Bank in 2021, Dr. Mitchell was professor of economic modeling and forecasting at the University of Warwick, where from 2013 to 2018 he headed the economic modeling and forecasting group at Warwick Business School. In academic year 2018–2019, he was a visiting scholar at the University of Southern California. Prior to Warwick, he was professor of economics and finance at the University of Leicester, and from 1999 to 2011 he worked at the National Institute of Economic and Social Research in London. He currently serves as an associate editor of the Journal of Applied Econometrics and the International Journal of Forecasting. In 2022, Dr. Mitchell was elected as a fellow of the International Association of Applied Econometrics.
Dr. Mitchell holds a BA in economics from Durham University, an MSc in economics from the University of Bristol, and a PhD in economics from the University of Cambridge.