A0540
Title: Heterogeneous panel data models with generalized cross-section dependence
Authors: John Goodhand - Air Liquide (United States) [presenting]
Abstract: A generalization of the asymptotic theory is presented for dynamic cross-section heterogeneous coefficient panels with interactive fixed effects, extending it to account for correlation among cross-section units due to local shocks. Traditional estimators in the multifactor error structure literature predominantly focus on the influence of global shocks on parameter estimates and inference. It goes beyond this narrow focus, incorporating the impact of local shocks that significantly affect only a small subset of the cross-section units in the sample. The limiting distribution for the cross-section heterogeneous coefficients is derived. Findings reveal a bias in the coefficient estimates related to local shocks associated with the weak cross-section dependence of the idiosyncratic error. Sufficient conditions for consistently estimating this bias and the covariance matrix of the limiting distribution are proposed in the presence of both local and global shocks. Theoretical findings are substantiated by Monte Carlo experiments, which demonstrate the superior finite sample performance of the estimation method over other competing techniques when the idiosyncratic errors are weakly cross-section dependent. Finally, an empirical application of the estimator is provided, evaluating the country-specific long-term impact of public and private debt on economic growth across 86 countries.
