On the stability of spatial econometric models: Application to the Czech Republic and its neighbors

Regional macroeconomic processes may not be properly analyzed without accounting for their spatial nature: regional distances, interactions between neighbors, spill-overs and interdependencies. Spatial econometrics is a unique tool for a broad range of quantitative analyses and evaluations with geo-coded (spatially defined) data. This contribution focuses on various factors of spatial model stability. Different spatial dependence matrices are used to evaluate the stability of the estimated unemployment dynamics in the Czech Republic, the Slovak Republic and all their EU-neighbors. The analysis is performed at the NUTS2 level. Spatial econometrics and regional competitiveness paradigms are combined into a robust model describing unemployment dynamics. Alternative spatial structures (i.e. neighbor definitions) are used for verification of stability of the estimated model and its properties. Spatial approach to econometric analysis provides important additional insight and robustness to a potentially broad range of spatially defined unemployment analyses that may be carried out using regional (spatial) data.