Being a former assistant professor from Laval University and a two-year F.R.S.-FNRS researcher from Université Namur, I am now working as an associate professor (full-time) at EDHEC Business School. My research mainly focuses on time series econometrics. In particular, I mostly use Bayesian inference to model macroeconomic and financial time series.
My Ph.D. supervisor was Luc Bauwens at Université Catholique de Louvain (UCL) and I was next a one-year postdoc at CREST as well as at UCL.
Current position (see CV) :
- Associate professor at EDHEC Bussiness school
Research Interests:
- Econometrics, Finance.
- Time series modeling and Bayesian inference.
- Time-varying parameter models.
Published papers :
- Dufays, A. and Li Zhuo and Rombouts, J. and Song Y., ‘Sparse Change-point VAR models‘, 2020. Accepted in Journal of Applied Econometrics.
- Dufays, A. and Houndetoungan, A. and Coën, A., ‘Selective linear segmentation for detecting relevant parameter changes‘, Accepted in Journal of Financial Econometrics, 2020.
- Donfack M. and Dufays A., ‘Modeling time-varying parameters using artificial neural networks: A GARCH illustration‘, accepted in Studies in Nonlinear Dynamics & Econometrics, 2020.
- Working paper [link]
- Dufays, A. and Rombouts, J., Relevant parameter changes in structural break models, Journal of Econometrics, 2020, 217(1), 46-78.
- Augustinyak, M. and Dufays A., Modeling macroeconomic series with regime-switching models characterized by a high-dimensional state space, Economic Letters, 2018, 170, 122-126.
- Augustinyak, M. and Bauwens, L. and Dufays A., A new approach to volatility modelling: the factorial hidden Markov volatility model, Journal of Business and Economic Statistics, 2018, 1-14.
- Dufays A. and Rombouts, J., Sparse Change-point HAR Models for Realized Variance, Econometric Reviews, 2018, 1-24.
- Bauwens, L., Carpantier J.-F., and Dufays A., Autoregressive Moving Average Autoregressive infinite Markov-switching models, Journal of Business and Economic Statistics, 2017, 35(2), 162-182.
- Working paper [link]
- Dufays, A., Evolutionary Sequential Monte Carlo Samplers for Change-Point Models, Econometrics, 2016, 4(1), 12.
- Dufays, A., Infinite-State Markov-Switching for Dynamic Volatility, Journal of Financial Econometrics, 2016, 14 (2), 418-460.
- Bauwens L., Dufays A. and De Backer B., A Bayesian method of Change-point estimation with recurrent regime : Application to GARCH Models, Journal of Empirical Finance, 2014, 29, 207-229.
- Working paper [link]
- Bauwens L., Dufays A. and Rombouts J., Marginal Likelihood Computation for Markov Switching and Change-point GARCH Models, Journal of Econometrics, 2013, 178 (3), 508-522.
Working papers and ongoing research :
- Dufays, A. and Jacobs, K. and Liu, Y. H. and Rombouts, J., ‘Fast Filtering with Large Option Panels: Implications for Asset Pricing’, 2021.
- Short presentation [link]
- Ardia, D. and Dufays, A. and Ordas, C., ‘Frequentist and Bayesian change-point models: a missing link’, 2020.
- Short presentation [link]
- Boucher V. and Dedewanou F. and Dufays A., ‘Peer-Induced Beliefs Regarding College Participation‘, 2018.
- Carpantier, J-F. and Dufays, A., ‘Commodities Volatility and the Theory of Storage‘, 2012.
Ph.D. thesis :
Modeling structural changes in volatility
(Thesis advisor : Luc Bauwens)
It is well known by economists that forecasting time series is difficult for many reasons, among which the changing (institutional, technological…) environment and behaviour of agents of the economic system. Changes may be slow and progressive or quick and abrupt, and generate the non-stationarity of many time series in the long run, such as changes in their trend or volatility. Ignoring these breaks by assuming constant parameters in econometric models, typically leads to forecasts that are far from realizations. The thesis focuses on detecting and estimating structural breaks in volatility of financial time series.