As economic crises periodically disrupt the economic activity, a large and continuously growing literature was dedicated to understanding the reasons behind the crises, their mechanism, effects and, most of all, the determinants of resilience capacity, and the ability to overcome hardships by adapting and changing. By preparing in advance for economic shocks through resilience building during good times, the impact of economic crises can be attenuated. Starting from these considerations, the paper focuses on regional economic specialization and its opposite – economic diversification, two business strategies already acknowledged in the literature as relevant factors for the capacity to mitigate economic crises. We tested the hypothesis of vulnerability-inducing economic specialization in the Romanian economy, using NUTS3 level data and found that more diversified regional economies were better at coping with the hardships triggered by the recent recession.
A common problem with using different statistical packages for the same data and method is the risk of getting dissimilar results. While the reasons behind this outcome are often known and accepted, the negative consequences might be significant. In a teaching environment, usually involving toy models, with no practical implications, only a reputation risk is at stake. Nevertheless, students should be aware of such incongruities, their causes and possible solutions. Starting from these considerations, our paper addresses the differences that arise between R and WarpPLS while applying the Partial Least Squares Path Modelling (PLS-PM) method. To this end we estimate a PLS-PM model for analysing health-positioning data, compare the results and explain how the two statistical packages differ and complement each other in an attempt to derive the best fit for the data.