Title: Estimating linear dynamic panel data models using nonlinear moment conditions
Authors: Markus Fritsch - University of Passau (Germany) [presenting]
Andrew Adrian Yu Pua - Xiamen University (China)
Joachim Schnurbus - University of Passau (Germany)
Abstract: GMM-estimation of linear dynamic panel data models based on nonlinear moment conditions that arise when there is no serial correlation in the error terms for balanced and unbalanced panels is implemented in R. Additionally, the theoretical properties of an IV-estimator based on these moment conditions are derived when both, cross-section and time series dimension are large. Efficiency gains are demonstrated in a Monte Carlo simulation and for a well-known household demand data set.