Michael Pfarrhofer (@mpfarrho) header
Michael Pfarrhofer (@mpfarrho) profile picture

AP @WU_econ, Bayesian econometrics and empirical macroeconomics (monetary, business cycles, now/forecasting)

Vienna, Austria
Joined June 2010
Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)Photo by Michael Pfarrhofer (@mpfarrho)
New working paper "A Bayesian Gaussian Process Dynamic Factor Model" with T. Chernis (BMA), N. Hauzenberger (Uni Strathclyde), and H. Mumtaz (QMUL): arxiv.org/abs/2509.04928 Proposes a DFM where the latent factors are linked to observed variables with unknown nonlinear functions
1
2
140
➡️ Nonparametric observation equation, specified via Gaussian process for each series ➡️ Factors modeled with a VAR (easy computation and interpretation) ➡️ Applications: (1) forecasting with FRED-QD, (2) extracting drivers of global inflation, measuring international asymmetries
57
#SUERFpolicybrief “Monetary policy under economic uncertainty: Evidence from the Euro Area" by Niko Hauzenberger (@UniStrathclyde), @mpfarrho (@wu_vienna), and Anna Stelzer (@oenb) 📄tinyurl.com/37re78uh
1
145
Michael Pfarrhofer retweeted
New CEPR Discussion Paper - DP20442 Nonparametric Mixed Frequency Monitoring Macro-at-Risk Massimiliano Marcellino @Unibocconi, @mpfarrho @wu_vienna @WU_econ ow.ly/q6GB50WmRMW #CEPR_MEF #EconTwitter
2
2
1,590
Michael Pfarrhofer retweeted
New CEPR Discussion Paper - DP19965 Bayesian Nowcasting with Mixed Frequency Data Using Gaussian Processes @NHauzenberger @Strath_Econ, Massimiliano Marcellino @Unibocconi, @mpfarrho @WU_econ, Anna Stelzer @oenb ow.ly/uO8S50V6S1v #CEPR_MEF #EconTwitter
1
1
2
1,632
"Scenario analysis with multivariate Bayesian machine learning models" ▶️ Conditional forecasts, generalized IRFs in dynamic nonparametric models ▶️ Macroeconomic stress testing; macroeconomic risk and financial conditions; financial shock transmission arxiv.org/abs/2502.08440
152
“Interpretable Bayesian machine learning for assessing the effects of climate news shocks on firm-level returns”: ▶️ Bayesian asset pricing framework & machine learning ▶️ text-based climate risk indicators ▶️ asymmetric impacts across firms ssrn.com/abstract=5133162
1
98
Michael Pfarrhofer retweeted
#Econ Job Vacancy — Post Doc, 2 years at WU Vienna Are you interested in collaborating with us (@WFrimmel , @HO2604 ) on research papers focused on intergenerational mobility in Austria? - Vienna University of Economics & Business (@wu_vienna) - Fulltime research position at the Department of Econ (@WU_econ) - Starting date in spring 2025 - Application Deadline: March 4, 2025 - More info: sites.google.com/view/alps-i… - Pls share/RT
9
1
11
7,669
Michael Pfarrhofer retweeted
You want to do your #PhD in #Economics at one of Europe’s largest and most modern business & economics universities. Join us @wu_vienna, @WU_econ. Click her econjobmarket.org/positions/… #Vienna 🇦🇹
2
10
867
Michael Pfarrhofer retweeted
WE ARE HIRING! Want to do a PhD in economics? Stop searching and click here: econjobmarket.org/positions/…
3
1
593
In this new (significantly revised) WP, @annstelzer and I (@WU_econ) discuss how to extract multiple structural shocks (monetary policy and central bank information) from multiple external instruments across multiple economies (US, EA and a few more) arxiv.org/abs/1912.03158
1
3
171
"Bayesian nonparametric methods for macroeconomic forecasting" (chapter 5), co-authored by Massimiliano Marcellino (@Unibocconi) and me (@WU_econ) #econometrics #forecasting doi.org/10.4337/978103531005…
📣 NEW: Handbook of Research Methods and Applications in Macroeconomic Forecasting by Michael P. Clements & Ana Beatriz Galvão Chapter 14 is #OpenAccess at: doi.org/10.4337/978103531005… More info: e-elgar.com/shop/isbn/978103… #MacroeconomicForecasting #Econometrics @AndrewBMartinez
1
4
359
Final few days left (deadline on November 20) to apply for the non-tenure track Assistant Professor position (post-doc) in Macroeconomics at @WU_econ at @wu_vienna #EconJobMarket econjobmarket.org/positions/…
See also on @EconJobMarket, for the macro AP position (deadline on Nov 20, 2024) @wu_vienna: econjobmarket.org/positions/…
1
182
See also on @EconJobMarket, for the macro AP position (deadline on Nov 20, 2024) @wu_vienna: econjobmarket.org/positions/…
WE ARE HIRING! We offer three positions as Assistant Professor, non-tenure track: IO, macroeconomics, and public economics. Details at wu.ac.at/karriere/jobportal/…
1
1
2
954
Michael Pfarrhofer retweeted
Welcome to our new colleagues: Jonathan Fitter, Nicolas Göller, Maximilian Heinze, Sannah Tijani (PhD program) & Kentaro Asai, Federica Braccioli (@fede_braccioli), Davide Cerruti (@dcerruti), Shahroo Malik, Bernhard Schmidpeter (@BernhardEcon), Nina Xue (Assistant Professors)!
4
27
3,002
“Sparse time-varying parameter VECMs with an application to modeling electricity prices” by N. Hauzenberger (@UniStrathclyde), L. Rossini (@LaStatale) and me (@wu_vienna, @WU_econ) published in the International Journal of Forecasting (IJF, @IIForecasters) sciencedirect.com/science/ar…
1
3
136
➡ Time-varying parameter (TVP) vector error correction model (VECM) with heteroskedastic errors ➡ Automatic dynamic model specification with cointegrated time series: global–local priors and postprocessing to achieve sparsity ➡ Applied to modeling European electricity prices
86
Michael Pfarrhofer retweeted
New research by @mpfarrho (@wu_vienna) and co-authors develops Bayesian machine learning tools, harnessing macroeconomic big data to accurately monitor the current state of the economy. arxiv.org/abs/2402.10574
1
2
2
413
"Nowcasting with Mixed Frequency Data Using Gaussian Processes" by N Hauzenberger (@UniStrathclyde), M Marcellino (@Unibocconi), A Stelzer (@oenb) and me (@WU_econ). We develop Bayesian machine learning methods for mixed data sampling (MIDAS) regressions. arxiv.org/abs/2402.10574
1
6
355
We evaluate predictions in short-horizon now- and forecasting exercises with both simulated data and data on quarterly US output growth and inflation in the GDP deflator.
1
67
Our proposed framework leverages macroeconomic Big Data in a computationally efficient way and offers gains in predictive accuracy compared to other machine learning approaches.
80