Exposiciones en situación de impago. Estimación de parámetros para el cálculo de capital regulatorio y su predicción mediante aprendizaje automático

Résumé

Regulatory developments on risk measurement are included across several documents published both by the European Banking Authority and the European Central Bank. On this basis, a concrete proposal of the Expected Loss Best Estimate (ELBE) and the Loss Given Default (LGD) in-default models is presented. The methodology is eventually calibrated based on data from the mortgage¿s portfolios of Spanish banks. The outcome serves to analyse the portfolios¿ risk profile. Recently, the economic onslaught of the COVID-19 pandemic compromised the financial risk management. Traditional methods fail to estimate the impact of such unprecedented situation. The ELBE is thus forecasted using a machine learning technique. The projection of two ELBEs for 2022 and their comparison are presented. One accounts for the outbreak¿s impact, and the other presumes its nonexistence. The proposed method has excellent performance and serve to estimate future losses amidst any event of extraordinary magnitude.