Yousef Kaddoura
Email (Main): yousef.kaddoura@liverpool.ac.uk
Other: yousef.kaddoura@nek.lu.se
Scholar: Google Scholar
Welcome!
I am an Assistant Professor at the University of Liverpool Management School, and I am also affiliated with Lund University School of Economics and Management.
I received my Ph.D. from Lund University in May 2024. Additionally, I hold an M.Sc. in Economics and an M.Sc. in Finance from Lund University.
Research interests: Econometrics, Machine learning, Panel Data, Large factor models, Regularization techniques, Clustering/Grouping, Structural breaks.
Latest publication:
Estimation of Panel Data Models with Random Interactive Effects and Multiple Structural Breaks when T is Fixed (with Joakim Westerlund, in Journal of Business and Economic Statistics)
In this article, we propose a new estimator of panel data models with random interactive effects and multiple structural breaks that is suitable when the number of time periods, T, is fixed and only the number of cross-sectional units, N, is large. This is done by viewing the determination of the breaks as a shrinkage problem, and to estimate both the regression coefficients, and the number of breaks and their locations by applying a version of the Lasso approach. We show that with probability approaching one the approach can correctly determine the number of breaks and the dates of these breaks, and that the estimator of the regime-specific regression coefficients is consistent and asymptotically normal. We also provide Monte Carlo results suggesting that the approach performs very well in small samples, and empirical results suggesting that while the coefficients of the controls are breaking, the coefficients of the main deterrence regressors in a model of crime are not.
Some hights of the paper and implications in economics:
1) In this paper, we address an issue that we believe is highly relevant in economics.
2) We estimate a panel data model that not only allows for interactive effects but also accommodates structural breaks.
3) Interactive effects are a generalization of fixed effects.
4) Structural breaks mean that we permit the coefficient estimates to change over time.
5) Failing to account for interactive effects or structural breaks can compromise inference.
6) We employ a machine learning technique to estimate this model in our study.
7) The assumptions are suitable for both micro and macro data, presenting significant potential for empirical economics research.
8) Given the method's generality and computational efficiency, we believe it can be used in numerous applications to uncover the dynamics of coefficient estimates while accounting for unobserved heterogeneity.