A Simultaneous System of Dynamic Spatial Stochastic Frontier Models

In this paper, we propose a simultaneous system of stochastic frontiers for dynamic panel data models with spatial dependence. Specifically, the model allows differentiating between productivity spillovers within a single production frontier and among different production processes. This model is particularly relevant in empirical studies considering multi-output and multi-input economic systems characterized by interrelated input-output linkages, spatial correlation, and temporal dynamics. As an example, an application to European OECD data for the agricultural and food and beverage industrial sectors is provided.