iCEBDA Seminar Series

29/9/2025

Kien C. Tran — University of Lethbridge

kien.tran@uleth.ca

Recent Development in Instrument-Free Approaches to Regression Models with Endogenous Regressors

3 p.m. Moscow time

Download video

Abstract: The main purpose of this paper is to provide alternative and recent developments in the instrument-free approaches to the endogeneity problem in linear regression models. Specifically, various consistent estimations of the model's parameters are discussed when no external information is available to construct valid instruments. An empirical application is provided to illustrate the usefulness of these approaches.

6/10/2025

Yulong Wang — Syracuse University

ywang402@syr.edu

Genuinely Robust Inference for Clustered Data

3 p.m. Moscow time

Download video

Abstract: Conventional cluster-robust inference can be invalid when data contain clusters of unignorably large size. We formalize this issue by deriving a necessary and sufficient condition for its validity, and show that this condition is frequently violated in practice: specifications from 77% of empirical research articles in American Economic Review and Econometrica during 2020–2021 appear not to meet it. To address this limitation, we propose a genuinely robust inference procedure based on a new cluster score boot-strap. We establish its validity and size control across broad classes of data-generating processes where conventional methods break down. Simulation studies corroborate our theoretical findings, and empirical applications illustrate that employing the proposed method can substantially alter conventional statistical conclusions.

17/10/2025

Abderrahim Taamouti — University of Liverpool

abderrahim.taamouti@liverpool.ac.uk

Systemic Growth-at-Risk and Growth Spread Measures

12 noon Moscow time

Download video

Abstract: This paper introduces two novel sets of forward-looking macroeconomic metrics-Systemic Growth-at-Risk (GaR) Measures and Growth Spread Measures-to quantify how growth risks propagate across countries and to assess the net growth benefits of regional economic integration, respectively. The Systemic GaR Measures capture the transmission of downside risks between countries, while the Growth Spread Measures evaluate asymmetric growth outcomes during periods of union-wide expansions and contractions. Applying these tools to the European Union (EU), we illustrate how integration simultaneously fosters shared growth potential and heightens exposure to systemic shocks. We estimate time-varying growth distributions for 18 OECD European countries using two complementary multivariate approaches: a GARCH-Copula framework with copula-based simulation, and a GARCH-Dynamic Conditional Correlation (DCC) model combined with a nonparametric bootstrap method. These techniques generate realistic joint growth scenarios that account for both idiosyncratic shocks and crosscountry interdependence. The results reveal substantial heterogeneity in the growth dividends of EU membership. Panel regressions attribute this variation to structural country characteristics such as trade openness, fiscal stance, development level, and exposure to global uncertainty. The framework extends the GaR literature into a multi-country setting, offering new insights into the dual nature of economic unions as both stabilizers and amplifiers of risk.

10/11/2025

Marco Barassi — University of Birmingham

m.r.barassi@bham.ac.uk

Threshold Regression in Heterogeneous Panel Data with Interactive Fixed Effects

12 noon Moscow time

Download video

Abstract: This paper introduces unit-specific heterogeneity in panel data threshold regression. We develop a comprehensive asymptotic theory for models with heterogeneous thresholds, heterogeneous slope coefficients, and interactive fixed effects. Our estimation methodology employs the Common Correlated Effects approach, which is able to handle heterogeneous coefficients while maintaining computational simplicity. We also propose a semi-homogeneous model with heterogeneous slopes but a common threshold, revealing novel mean group estimator convergence rates due to the interaction of heterogeneity with the shrinking threshold assumption. Tests for linearity are provided, and also a modified information criterion which can choose between the fully heterogeneous and the semi-homogeneous models. Monte Carlo simulations demonstrate the good performance of the new methods in small samples. The new theory is applied to examine the Feldstein-Horioka puzzle and it is found that threshold nonlinearity with respect to trade openness exists only in a small subset of countries.

1/12/2025

Yiannis Karavias — Brunel University of London

yiannis.karavias@brunel.ac.uk

Interactive, Grouped and Non-separable Fixed Effects: A Practitioner's Guide to the New Panel Data Econometrics

12 noon Moscow time

Download video

Abstract:  The past 20 years have brought fundamental advances in modeling unobserved heterogeneity in panel data. Interactive Fixed Effects (IFE) proved to be a foundational framework,generalizing the standard one-way and two-way fixed effects models by allowing the unitspecific unobserved heterogeneity to be interacted with unobserved time-varying commonfactors, allowing for more general forms of omitted variables. The IFE framework laid thetheoretical foundations for other forms of heterogeneity, such as grouped fixed effects (GFE)and non-separable two-way fixed effects (NSTW). The existence of IFE, GFE, or NSTWhas significant implications for identification, estimation, and inference, leading to the development of many new estimators for panel data models. This paper provides an accessiblereview of the new estimation methods and their associated diagnostic tests and offers a guideto empirical practice. In two separate empirical investigations we demonstrate that there isempirical support for the new forms of fixed effects and that the results can differ significantly from those obtained using traditional fixed effects estimators.

16/02/2026

Weifeng Jin — Instituto Tecnológico Autónomo de México (ITAM)

weifeng.jin@itam.mx

Estimation of Time Series Models Using the Empirical Distribution of Residuals

5 p.m. Moscow time

Download video

Abstract:  Nonfundamental representations of univariate processes have been applied in the fields of Macroeconomics and Finance to describe nonlinear dynamics resulting from future shocks. This paper introduces a novel estimation technique for general linear time series models, potentially noninvertible and noncausal, by utilizing the empirical cumulative distribution function of residuals. The proposed method relies on the generalized spectral cumulative function to characterize the pairwise dependence of the residuals at all lags. Model identification can be achieved by exploiting the information in the joint distribution of residuals under the iid assumption. This method yields consistent estimates of the model parameters without imposing stringent conditions on the higher-order moments or any distributional assumptions on the innovations beyond non-Gaussianity. We investigate the asymptotic distribution of the estimates by employing a smoothed cumulative distribution function to approximate the indicator function, considering the non-differentiability of the original loss function. Efficiency improvements can be achieved by properly choosing the scaling parameter for residuals. Finite sample properties are explored through Monte Carlo simulations. An empirical application to illustrate this methodology is provided by fitting the daily trading volume of Microsoft stock by autoregressive models with noncausal representation. The flexibility of the cumulative distribution function permits the proposed method to be extended to more general dependence structures where innovations are only conditional mean or quantile independent.

16/03/2026

Fabrizio Ghezzi — University of California San Diego

fghezzi@ucsd.edu  

TBD

Zoom link

20/04/2026

Valentina Corradi — NYU Abu Dhabi

vc2718@nyu.edu  

TBC

Zoom link

18/05/2026

Luca Fanelli — University of Bologna

 luca.fanelli@unibo.it

Exogeneity in Structural Vector Autoregressions and Local Projections with External Instruments

Zoom link