Using generalized estimating equations to estimate nonlinear models with spatial data

We study the estimation of nonlinear models with cross-sectional data using two-step generalized estimating equations in the quasi-maximum likelihood estimation framework. To improve efficiency, we propose a grouped estimator to account for the potential spatial correlation in the underlying innovations for nonlinear models. Under mild weak dependence assumptions, results on estimation consistency and asymptotic normality are provided. Monte Carlo simulations show the efficiency gain of our approach in comparison with different estimation methods. Finally, we apply the proposed approach to study the role of cultural distance in an extended gravity equation with the international trade data of China.