Cookie usage policy

The website of the University Carlos III of Madrid use its own cookies and third-party cookies to improve our services by analyzing their browsing habits. By continuing navigation, we understand that it accepts our cookie policy. "Usage rules"

Fabio Canova


Fabio Canova - Norwegian Business School. Norway

Fabio Canova is a professor of Macroeconomics at the Norwegian Business School, Research associate with the Centre for Applied Macroeconomics and Petroleum Studies
and the CEPR. He is also program director of the Budapest School of Central Bank Studies, and member of the scientific committee of the Euro Area Business Cycle network.
He was the director of Training of the Florence School of Banking and Finance (2015-2018) held the Pierre Werner chair in Monetary Union at the Robert Schumann Center for Advanced Studies (2012-2014), the ICREA Research Professorship at Universitat Pompeu Fabra (2006-2012) and has been Professor of Econometrics at the European University Institute (2011-2014) and Chair in Monetary Economics and the University of Bern (2008). In 2017 he was awarded a honorary professorship from Henin University in China.

In the recent past he has been program committee member of the meetings of the International Association of Applied Econometrics (2014-2017) and chair in 2020; chair for the European Meetings of the Econometric Society 2014, a panelist of ANVUR in 2013, coeditor of the Journal of the European Economic Association from 2008 to 2013, and of the Journal of Applied econometrics from 2012 to 2017.


Project: FAQs: Which filter should I use to extract the output gap? Are gaps fluctuations cyclical? Are gaps and transitory fluctuations the same?

I employ the Smets and Wouter (2007)  model as a lab to produce time series for potentials and gaps, and for permanent and transitory fluctuations in nine endogenous macroeconomic time series. Gaps have important low frequency variations, regardless of the calibration of the generating economy; display more than cyclical fluctuations, have similar frequency domain representation as potentials, and are correlated with them. These features are robust to adding financial frictions to the model or eliminating certain shocks. Away from the zero frequency, permanent and transitory fluctuations also have similar features, but are uncorrelated. Gaps are different from transitory fluctuations in a number of dimensions.
I use a variety of filters on the simulated level data and ask which procedure produce filtered series which look like the model' gaps, or the model's transitory fluctuations. While distortions are large, gaps are best approximated with a polynomial filtering approach. Transitory fluctuations are best recovered with a simple first order filter. Explanations for the results will be given.