Join us for this weeks Econometrics Research Seminar, Autumn Term 2024
Pavel Cizek, from Tilburg University, will present their research on Robust Indirect Inference Estimation of Linear Dynamic Panel Data Models.
Abstract
We introduce new estimation methodology for linear dynamic panel-data models. The proposed approach generalizes the estimation methods based on bias-corrected least-squares estimators to the robust regression estimators, thus providing an alternative that is robust to outliers and observations not following the specified dynamic model. For this purpose, we extend the existing simulation-based indirect-inference estimation, and to achieve sufficient robustness, we introduce the multivariate-trimmed estimators. We provide examples of the multivariate-trimmed estimators relevant to dynamic panel-data models, establish their asymptotic properties, and correct their biases using the proposed bias-correction method based on the indirect inference. The finite sample performance – both in terms of precision and robustness – is compared with existing estimators by means of Monte Carlo simulations, and it is demonstrated in a real-data application.
This seminar will be held on campus, is open to all levels of study and is also open to the public. To register your place and gain access to the webinar, please contact the seminar organisers.
This event is part of the Econometrics Research Seminar Series.