Drivers of scientific knowledge production. A regional perspective

The Focus of this master Thesis is on regional knowledge production activities in Europe, laying - in contrast to previous empirical works - Special emphasis on scientific knowledge producation as captured by publications of articles in scientific journals. The objective is to estimate to what extent different regional characteristics influence scientific knowledge production activities at the Level of European regions. A Bayesian hierarchical model is implemented that controls for both space and time aspect of the data. The use of such a model is a novelty as it is often employed in environmental sciences but not economics. The European coverage is achieved using 238 NUTS-2 regions covering 23 member states. The dependent variable, scientific knowledge production activity at the regional level, is measured in Terms of publication counts. The independent variables include tertiary education attainment, R&D expenditures, networking intensity, student density as well as a proxy for a high-ranked university presence in a region. Networking intensity, student concentration and presence of a high-ranked university prove to habe strong, positive effects on the scientific knowledge generation. Basing on These results a projection of future publication counts has been calculated. A positive trend has been identified with a tendency to slow down over the projection horizon.