ICEBDA Research Seminar Marco Barassi (University of Birmingham): “Threshold Regression in Heterogeneous Panel Data with Interactive Fixed Effects”
Centre for Big Data in Economics and Finance of the Faculty of Economic Sciences at HSE University invites you to participate in the research seminar: “Threshold Regression in Heterogeneous Panel Data with Interactive Fixed Effects” with Marco Barassi (University of Birmingham) on November 10, 2025 at 12:00 (Moscow time).
The Centre for Big Data in Economics and Finance at the HSE University Faculty of Economic Sciences invites you to the regular research seminar, which will be held on November 10, 2025, at 12:00 noon (Moscow time) online via Zoom.
Speaker:
Marco Barassi — University of Birmingham (UK)
Topic: “Threshold Regression in Heterogeneous Panel Data with Interactive Fixed Effects”
Contact: m.r.barassi@bham.ac.uk
The seminar will present research that introduces unit-specific heterogeneity into panel data threshold regression. The speaker has developed a comprehensive asymptotic theory for models featuring heterogeneous thresholds, heterogeneous slope coefficients, and interactive fixed effects. The proposed estimation methodology employs the Common Correlated Effects approach, which handles heterogeneous coefficients while maintaining computational simplicity. A semi-homogeneous model with heterogeneous slopes but a common threshold will also be presented, revealing novel convergence rates for mean group estimators. The study includes tests for linearity and a modified information criterion for selecting between the fully heterogeneous and semi-homogeneous models. The methods demonstrate strong performance in small samples via Monte Carlo simulations. Applying the new theory to the Feldstein-Horioka puzzle, the author finds that threshold nonlinearity with respect to trade openness exists only in a small subset of countries.
The Centre for Big Data in Economics and Finance, established in 2025 at the HSE University Faculty of Economic Sciences, conducts fundamental and applied research in big data analysis, develops modern econometric and machine learning methods, and implements educational programs aimed at training the next generation of specialists. Learn more about the Centre's activities on its official webpage.
To participate, please register on the Centre's seminar page, where the Zoom link will be provided upon confirmation.