Univ.-Prof.Dipl.-Ing.MMMag.Dr. Gregor Kastner

E-Mail: Gregor [dot] Kastner [at] aau [dot] at
Telefon: +43 463 2700 3440

Raum:

N.0.08a Hauptgebäude, Nordtrakt West, Ebene 0 (Institut für Statistik)

Zugehörigkeit:

Institut für Statistik

Campus-Portal Visitenkarte | Lehrveranstaltungen | Forschung

Forschungsschwerpunkte / ÖSTAT-Sachgebiete

Ökonometrie
Data Science
Softwareentwicklung
Statistik
Zeitreihenanalyse

Curriculum Vitae

Laufbahn

I am interested in Bayesian modelling of time series data, mainly through Markov chain Monte Carlo (MCMC) methods. This usually requires high performance computing infrastructures. My current research focus is the efficient Bayesian estimation of univariate and multivariate stochastic volatility (SV) and vector autoregressive (VAR) models as well as statistical modeling of economic data. I am founder of and active contributor to the open-source R packages stochvol and factorstochvol.

Ausbildung

2014: Dr.techn. (JKU Linz, WU Wien: Applied Mathematics)
2011: Mag.rer.nat. (University of Vienna: Physical Education)
2009: Mag.rer.nat. (TU Wien, University of Vienna: Computer Science)
2009: Mag.rer.soc.oec. (TU Wien, University of Vienna: Informatics Management)
2006: Dipl.-Ing. (TU Wien, ETH Zurich: Applied Mathematics)

Curriculum Vitae

Ausgewählte Projekte

Ausgewählte Publikationen

2025
Luis Bastian Gruber, Gregor Kastner, Anirban Bhattacharya, Debdeep Pati, Natesh Pillai, David Dunson: A Note on Simulation Methods for the Dirichlet-Laplace Prior. Journal of the American Statistical Association, Taylor & Francis, 120, 2025, S. 2011 - 2014.
Laura Vana-Gür, Ennio Visconti, Laura Nenzi, Annalisa Cadonna, Gregor Kastner: Bayesian Machine Learning Meets Formal Methods: An Application to Spatio-Temporal Data. ACM Transactions on Probabilistic Machine Learning, 1, 2025, S. 1 - 30.
Luis Bastian Gruber, Gregor Kastner: Forecasting macroeconomic data with Bayesian VARs: Sparse or dense? It depends!. International Journal of Forecasting (IJF), Elsevier, 41, 2025,
2024
Martin Feldkircher, Luis Bastian Gruber, Florian Huber, Gregor Kastner: Sophisticated and small versus simple and sizeable: When does it pay off to introduce drifting coefficients in Bayesian vector autoregressions?. Journal of Forecasting, John Wiley & Sons. Ltd., 2024,
2022
Alexander Mozdzen, Cremaschi Andrea, Annalisa Cadonna, Guglielmi Alessandra, Gregor Kastner: Bayesian modeling and clustering for spatio-temporal areal data: An application to Italian unemployment. Spatial Statistics, Elsevier B.V., 52, Amsterdam, 2022, S. 100715
2021
Darjus Hosszejni, Gregor Kastner: Modeling Univariate and Multivariate Stochastic Volatility in R with stochvol and factorstochvol. Journal of Statistical Software, 100, 2021, S. 1 - 34.

Ausgewählte Vorträge

2025
High-Dimensional Statistical Learning for (Financial and Economic) Time Series: A Bayesian's View
Forschungsseminar
Universität zu Köln, 10.06.2025
Sparse Dynamic Bayesian Graphical Models
BAYSM 2025
online , 08.04.2025
2024
Sparse dynamic Bayesian graphical models
18th International Joint Conference CFE-CMStatistics
London, 16.12.2024
Sparse Dynamic Bayesian Graphical Models
14th European Seminar on Bayesian Econometrics (ESOBE 2024)
Örebro, 22.08.2024
Bayesian Machine Learning meets Formal Methods: An application to spatio-temporal data
BNP Networking Workshop 2024
Singapur, 01.08.2024
Dynamic Sparsity in Factor Stochastic Volatility Models
2024 ISBA World Meeting
Venice, 02.07.2024
Forecasting Macroeconomic Data With Bayesian Vars: Sparse Or Dense? It Depends!
44th International Symposium on Forecasting
Dijon, 01.07.2024
2023
Forecasting macroeconomic data with Bayesian VARs: Sparse or dense? It depends!
BAYSM 2023
Online, 15.11.2023
Time-varying Risk Premia and Volatility Dynamics in Multi-Asset Class Returns
13th European Seminar on Bayesian Econometrics
Glasgow, 01.09.2023
Posterior predictive model assessment using formal methods in a spatio-temporal model
Bayes Comp 2023
Levi, Finland, 16.03.2023
2022
Bayesian Methods for (Macro and Financial) Time Series Analysis
3rd Klagenfurt-Bielefeld Summer School on Modern Topics in Time Series Analysis 2022
Klagenfurt, 14.09.2022
Time-varying Risk Premia and Volatility Dynamics in Multi-Asset Class Returns
5th Vienna Workshop on High-Dimensional Times Series in Macroeconomics and Finance
Wien, 10.06.2022
Modelling Stock-Bond Correlations
VieCo 2022
Kopenhagen, 02.06.2022
Time Series Predictions and Model Comparison in the Bayesian Paradigm
Statistical Days 2022
Graz, 22.04.2022
Posterior predictive model checking using formal methods in a spatio-temporal model
ISOR Colloqium
Wien, 10.01.2022
2021
Posterior predictive checking and verification using formal methods in a spatio-temporal model
DMV-ÖMG Annual Conference 2021
Passau, 27.09.2021
2020
Time-varying Risk Premia and Volatility Dynamics in Multi-Asset Class Returns
Bayes@Austria 2020
Wien, 28.11.2020
Efficient Bayesian Inference in High Dimensional Time Series : Examples from Macroeconomics and Finance
Séminaires Dagenais économétrie
Montréal, 26.11.2020

Wissenschaftliche Funktionen