This paper presents a discussion on several econometric models for estimating parametric models in the presence of spatial dependence with cross-section data. Initially, we cover models for spatial dependence with spatial lags of the response variable and spatial lags of the residues, and estimation is accomplished by maximum likelihood. A critical analysis for these models is also presented, followed by a discussion on tests for spatial dependence. Finally, we present a discussion no more robust estimation methods, allowing for endogeneity in some of the explanatory variables.