Model uncertainty, nonlinearities and out-of-sample comparison: evidence from international technology diffusion

This paper reconsiders the international technology diffusion model. Because the high degree of uncertainty surrounding the Data Generating Process and the likely presence of nonlinearities and latent common factors, it considers alternative nonparametric panel specifications which extend the Common Correlated Effects approach and then contrasts the out-of-sample performance of them with those of more common parametric models. To do so, we adopt an approach recently proposed within the literature of nonparametric regression. This approach is based on a pseudo Monte Carlo experiment that takes its roots on cross validation and aims at testing whether two competing approximate models are equivalent in terms of their expected true error. Our results indicate that the adoption of a nonparametric approach provides better performances. This work also refines previous results by showing threshold effects, nonlinearities and interactions, which are obscured in parametric specifications and which have relevant implications for policy.