| Abstract [eng] |
The thesis consists of two main parts. In the first part, the aim was to apply econometric models that help to establish links between the dynamics of macroeconomic variables and the developments in banks' ratios. These relationships are important assessing in advance the resilience of the banking system to various risks that may arise in the near future. A new top-down stress testing framework was introduced. This framework was used assessing the resilience of the banking system to an adverse economic development. Furthermore, a panel error correction model was employed to assess long-term and short-term internal and external determinants of items from the bank income statement. In the second part of the work, the clustering of a given set of the European banks was discussed and the similarity/dissimilarity between banks was estimated, using measures based on time series or functional data properties. When making the cluster analysis, two dissimilarity measures, not commonly used in the literature, were proposed. Finally, the advantages of the functional data analysis methods were explored and applied to the extracted clusters. The theoretical functional data models were proposed and afterwards used for the top-down and bottom-up forecasting problem. |
