A0466
Title: Dynamic quantile panel data models with interactive effects
Authors: Chaowen Zheng - University of Southampton (United Kingdom) [presenting]
Abstract: A simple two-step procedure is proposed for estimating the dynamic quantile panel data model with unobserved interactive effects. Factors are first consistently estimated via an iterative principal component analysis to account for the endogeneity induced by the correlation between factors and lagged dependent variable/regressors. In the second step, a quantile regression is run for the augmented model with estimated factors and estimated slope parameters. In particular, a smoothed quantile regression analysis is adopted where the quantile loss function is smoothed to have well-defined derivatives. The proposed two-step estimator is consistent and asymptotically normally distributed but subject to asymptotic bias due to the incidental parameters. The split-panel jackknife approach is then applied to correct the bias. Monte Carlo simulations confirm that the proposed estimator has good finite sample performance. Finally, the usefulness of the proposed approach is demonstrated with an application to the analysis of bilateral trade for 380 country pairs over 59 years.
