An Adaptive Neural Network Approach To Predict The Capital Adequacy Ratio
| Autor: | Giacomo Di Tollo, Gerarda Fattoruso, Bartolomeo Toffano |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2022 |
| Předmět: | |
| Zdroj: | Ratio Mathematica, Vol 43, Iss 0, Pp 183-207 (2022) |
| Druh dokumentu: | article |
| ISSN: | 1592-7415 2282-8214 |
| DOI: | 10.23755/rm.v43i0.841 |
| Popis: | Financial institutions, policy makers and regulatory authorities need to implement stress tests in order to test both resilience and the consequences of adverse shocks. The European Central Bank and the European Banking Authority regularly conduct these tests, whose importance is more and more evident after the financial crisis of 2007-2008. The stress tests’ nonlinear features of variables and scenarios triggered the need of general and robust strategies to perform this task. In this paper we want to introduce an adaptive Neural Network approach to predict the Capital Adequacy Ratio (CAR), which is one of the main ratios monitored to retrieve useful information along many stress test procedures. The Neural Network approach is based on a comparison between feed-forward and recurrent networks, and is run after a meaningful pre-processing operations definition. Results show that our approach is able to successfully predict CAR by using both Neural Networks and recurrent networks. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|