I am a Senior Lecturer in the Department of Economics at the University of Strathclyde and Director of the MSc programme in Economics, Data Science and AI.
My research focuses on developing novel econometric methods for the efficient use of big data in macroeconomics. In particular, I combine modeling techniques from the machine learning and Bayesian learning literature with multivariate time series models that macroeconomists commonly work with (e.g., vector autoregressions). Applications include the transmission of monetary policy and uncertainty shocks, inflation dynamics, forecasting and nowcasting, scenario analysis, and macro-financial linkages. My work has been published in the Journal of Econometrics, the Journal of Business & Economic Statistics, the International Economic Review, the Journal of Applied Econometrics, the International Journal of Forecasting, among others.
Selected (recently published) papers.
- Direct Gaussian process predictive regressions with mixed frequency data, with M. Marcellino, M. Pfarrhofer, and A. Stelzer. Journal of Business & Economic Statistics, forthcoming.
- Bayesian modeling of time-varying parameter vector autoregressions using regression trees, with F. Huber, G. Koop, and J. Mitchell. Annals of Applied Statistics, 2026.
- Bayesian neural networks for macroeconomic analysis, with F. Huber, K. Klieber, and M. Marcellino. Journal of Econometrics, 2025.
Current work.
- Conditional projection methods for large-scale Bayesian VARs, with M. Pfarrhofer. Fast conditional forecasting and scenario analysis in high-dimensional VARs, applied to oil price scenarios after the 2026 closure of the Strait of Hormuz.
- A Bayesian Gaussian process dynamic factor model, with T. Chernis, H. Mumtaz, and M. Pfarrhofer. A factor model with a nonparametric observation equation, used to forecast US macro series and to extract drivers of global inflation.
I am first supervisor of three PhD students. Two of them work on funded studentship projects that I lead as principal investigator. The project Central Banks’ inflation gamble: Good luck vs. good models (with G. Koop and S. McIntyre) uses machine learning methods to model and forecast inflation at the level of its subcomponents and not only the aggregate. The aim is to understand what drove the recent inflation surge and to contribute to the good luck versus good policy debate. The project Addressing the economic data crisis: Bayesian statistical learning for real-time forecasting and decision support (with C. Bueno Macedo Medeiros and G. Koop) deals with real-time macroeconomic forecasting when official data arrive late, get revised, or are missing. It combines Bayesian machine learning with data at different frequencies and from less conventional sources, such as surveys and text, and also measures tail risks to the economy.
My research typically involves close collaborations with researchers and economists at policy institutions and central banks, including the Austrian Central Bank (OeNB), the Bank of Canada, the Federal Reserve Bank of Cleveland, the International Institute for Applied Systems Analysis (IIASA), and the Joint Research Centre (JRC) of the European Commission. Active knowledge exchange is part of this work. I was a consultant to the OeNB and a research scholar at IIASA, worked with the Macro Modelling and Nowcasting Team at the JRC, and taught a continuing professional development course at the Banco de Portugal.
Background. Senior Lecturer, University of Strathclyde (since 2023). Postdoctoral researcher, University of Salzburg (2020–23), and postdoctoral fellow, WU Vienna (2022). PhD in Economics, WU Vienna (2020). Würdigungspreis of the Austrian Federal Ministry of Education, Science and Research (2018).