Francesco Audrino

Francesco Audrino is a Professor of Statistics at the Department of Mathematics and Statistics, University of St. Gallen, affiliated with the School of Economics and Political Science (SEPS). He holds a Diploma in Mathematics from ETH Zurich (specializing in Financial and Insurance Mathematics) and a Ph.D. in Statistics/Finance from ETH Zurich, with a thesis on statistical methods for high-dimensional financial time series.

Research Focus: Computational Statistics applied to Economics and Finance, Financial Econometrics, Sentiment Analysis, Volatility Estimation, and Regime-switching Models. His work emphasizes nonparametric methods (e.g., Functional Gradient Descent) and their applications in asset pricing, interest rate dynamics, and risk management. Recent projects include the SentiVol program for sentiment-driven volatility forecasting and causal machine learning analyses of post-earnings sentiment impacts.

• Selected Projects:
• Director of the SentiVol SNF Grant (2017–2020) on sentiment analysis for volatility prediction.
• Co-Director of the SNF Grant on Behavioral Asset Pricing (2010–2013).
• Leading research on cross-asset dependency structures and causal machine learning applications.

Awards & Recognition:

• Swiss Society of Economics and Statistics Young Economist Award (2007).
• University of St. Gallen’s Best Researcher Award (2007).
• Teaching load reduction for outstanding publications (2009–2018).

Teaching & Academic Service: Teaches Statistics, Financial Volatility, and Computational Statistics at bachelor’s, master’s, and Ph.D. levels. Served as Elected Member of the Board of Directors of the European Regional Section of the International Association for Statistical Computing (2016–2020). Active in academic networks like the Computational Financial Econometrics (CFE) Network.

Labs & Resources: Developed the FGD Code package for nonparametric high-dimensional time series analysis, widely used in volatility and correlation modeling. Maintains real-time volatility forecasts for S&P 500 constituents via the SentiVol project.