Wansbeek and Bekker considered a new estimator for simple dynamic panel data models (where there are no exogenous variables) which involved a complex weighting matrix. An operational variant of this estimator is proposed which is applicable to the more realistic case where there are exogenous...

In this paper Kuznet's U-curve hypothesis is tested on two unbalanced panel data sets of 47 and 62 countries, for the period 1970-93, using two-way fixed and random effects models. Several competing model specifications are estimated and the one best fitting the data is selected by appropriate...

Firm-level innovation is investigated using three probit panel estimators, which control for unobserved heterogeneity, and a standard probit estimator. Results indicate the standard probit model is misspecified and that inter-firm networks are important for innovation.