Nonstationary panel models with latent group structures and cross-section dependence

This paper proposes a novel Lasso-based approach to handle unobserved parameter heterogeneity and cross-section dependence in nonstationary panel models. In particular, penalized principal component (PPC) method is developed estimate group-specific long-run relationships common factors jointly identify the unknown group membership. The PPC estimators are shown be consistent under weakly dependent innovation processes. But they suffer an asymptotically non-negligible bias from correlations between regressors stationary and/or equation errors. To remedy these shortcomings we provide three bias-correction procedures which re-centered about zero as both dimensions (N T) of tend infinity. We establish mixed normal limit theory for coefficients, permits inference using standard test statistics. Simulations suggest good finite sample performance. An empirical application applies methodology study international R&D spillovers results offer convincing explanation growth convergence puzzle through heterogeneous impact spillovers.