This paper considers a class of generalized methods of moments (GMM) estimators for general dynamic panel models, allowing for weakly exogenous covariates and cross‐sectional dependence due to spatial lags, unspecified common shocks, and time‐varying interactive effects. We significantly expand the scope of the existing literature by allowing for endogenous time‐varying spatial weight matrices without imposing explicit structural assumptions on how the weights are formed. An important area of application is in social interaction and network models where our specification can accommodate data dependent network formation. We consider an exemplary social interaction model and show how identification of the interaction parameters is achieved through a combination of linear and quadratic moment conditions. For the general setup we develop an orthogonal forward differencing transformation to aid in the estimation of factor components while maintaining orthogonality of moment conditions. This is an important ingredient to a tractable asymptotic distribution of our estimators. In general, the asymptotic distribution of our estimators is found to be mixed normal due to random norming. However, the asymptotic distribution of our test statistics is still chi‐square.

Uniform laws of large numbers (ULLNs) consider sums of the form: n −1 Σ t n =1 [q t (z t , θ)-Eq t (z t , θ)], where (z t ) denotes a stochastic data generating process that takes its values in a space Z, θ is an element of the parameter space Θ, and q t : Z×Θ→R. ULLNs provide conditions under which the above sum converges to zero uniformly over the parameter space. The purpose of the present note is to introduce a new generic ULLN. It maintains a set of assumptions that is relatively easy to verify and allows at the same time the analysis of a wide variety of estimators and models of interest in economics

A general linear simultaneous equation system with a multivariate Student t disturbance vector is considered. The normal equations of the corresponding maximum likelihood estimator are used as estimator generating equations to introduce a new class of estimators. Properties of large subclasses of these estimators are determined for disturbance vectors other than the multivariate Student t.