Innovación e comercio: Estimación dun modelo dinámico de datos de panel con coeficientes heteroxéneos

The objective of this paper is to estimate a dynamic panel data model with heterogeneous coefficients. Since usual “within” estimators do not have good properties in an autoregressive model, we introduce instrumental variables for a fixed effects model, by adapting the method used by Balestra and Nerlove (1966) for a model with random effects. This methodology permits us to identify the importance of sectorial and national effects in the relation between innovation and international trade, based on the existence of different sectorial patterns in the innovation process, as well as in the characteristics of National Systems of Innovation which influence the effect of technology on the configuration of trade patterns.