BREAKMETRICS project funded by Marie Curie Grant

The general objective of this proposal is to improve economic forecasts with the help of evolutionary econometric models, i.e. models that adapt to abrupt (or structural) changes in the economic environment, also called structural break or change-point models. The need for adaptive modelling is obvious in the light of the current economic conditions.

Economic forecasting is essential for decision making with respect to fiscal and monetary policy, public spending, and investment. For example, the monetary transmission mechanism has long and uncertain lags. Therefore, monetary policy should be forward-looking. To effectively ensure price stability, the central bank for example needs to make forecasts about the evolution of prices and output among others. When a structural break happens, conventional forecasting models can produce severe forecast errors since the forecasts do not comply with the new environment as given by the new data.


Conferences and workshops

Event with Allan Timmerman – March 2014


ESSEC Business school (Paris-Singapore) has a strong reputation in forecasting and econometrics. On March 31, 2014, in the framework of their research activities professors from the Econometrics & Statistics Group invited Professor Allan Timmerman from UCSD to give a presentation at ESSEC Business School and meet with Paris-based researchers in econometric forecasting. Professor Timmermann is one of the leading scholars in forecasting, has cutting edge contributions in Finance, Economics and Statistics. He is regularly invited by the IIF as a keynote speaker.

The topic of his seminar was about forecasting bond returns. Since the recent global financial crisis, monetary policy and the behaviour of investors have made bond markets an exciting research topic. While advanced techniques exist for modelling and forecasting stock returns, bond returns are typically forecast with simple linear models using variables such as the term spread and factors extracted from macro-economic variables. Several existing papers show in-sample excess return predictability. However, Timmermann argues that there is hardly any evidence that this predictability translates in real time sizeable out-of-sample economic gains. Applying Bayesian estimation techniques to a three variable model with stochastic volatility, nesting most of the existing literature, he shows that ignoring parameter and model uncertainty is very costly when it comes to forecasting bond returns. In fact, by combining several forecasting models significant economic gains are obtained. Another and somewhat counterintuitive new finding is that bond return predictability is higher during recessions than expansions.

European seminar on Bayesian Econometrics – November 2014


ESOBE stands for European Seminar on Bayesian Econometrics. This series of seminars was launched in 2010 with the first meeting in Rotterdam. In recent decades Bayesian econometrics has expanded enormously in areas such as optimal processing of information from different sources, efficient forecasting using sets of models, and measuring policy effectiveness and its associated risk.

The computational revolution in simulation techniques is a key ingredient in this expansion. Empirical applications of Bayesian econometrics deal with issues such as: risk management in international and financial markets, the education effect on income and individual entrepreneurship, measurement of policy effectiveness in the macro and monetary economy and individual decision making in marketing.

The ESOBE meetings have no particular theme but are intended as a discussion forum for new and recent research. Their aim is to bring together researchers and professionals interested in the application of Bayesian inference in economics in relatively small annual workshops that usually take two days.

The conference website can be found here

Empirical Finance Workshop – March 2015


This workshop brings together researchers interested in empirical finance in general and in modelling of risk premia in particular. A central topic in finance is the identification of risk premia, which is the extra payment that risk adverse investors require for bearing different risks, or rather, the market price attached to different sources of risk. While the equity return premium has been studied for decades, the recent literature focuses on the identification and estimation of premia related to variance, skewness, correlation, etc.

The conference website can be found here

ICT & Innovation Forecasting Workshops – May 2015


The dynamic, ever changing nature of the Information & Communications Technology (ICT) Industry is a challenge for business planners and forecasters. The rise of Twitter and the sudden demise of Blackberry are dramatic examples of the uncertainties of the industry; these events clearly demonstrate how radically the environment can change. Similarly, predicting demand, market penetration, new markets, and the impact of new innovations in the ICT sector offer a challenge to businesses and policymakers. This Workshop will focus on forecasting new services and innovation in this sector as well as the theory and practice of forecasting in the sector (Telcos, IT providers, OTTs, manufacturers).

The ICT & Innovation Forecasting Workshops are intended as a forum for discussion of new and recent research and application in this field. Their aim is to bring together researchers and professionals interested in the application of forecasting methods in ICT, telecommunications and innovation in a relatively small two days annual workshop.

The conference website can be found here

Advances in Time Series and Forecasting – November 2015


This is a time series conference in honour of Professor Jean-Pierre Indjehagopian who has an extraordinary career at ESSEC Business School. The purpose is to bring together leading scientists in the field of time series and forecasting to advance in terms of methodology and applications.

The conference website can be found here

Conference on Systemic Risk – Singapore – December 2015


The objective of the workshop is to enhance the understanding and quantitative measurement of systemic risk.

The conference website can be found here


  • Luc Bauwens (CORE – Universite Catholic de Louvain)
  • Arnaud Dufays (Laval Universtiy)
  • Aurore Delaigle (Melbourne University)
  • Lars Stentoft (University of Western Ontario)
  • Allan Timmermann (UCSD)
  • Abderrahim Taamouti (Durham Business School)
  • Francesco Violante (Aarhus Unversity)
  • Ingrid Van Keilegom (ISBA¬†– Universite Catholic de Louvain)