Abstract
The early prediction of bad debtors in revolving loans in Mexico is a relevant current issue. The proposed econometric model of behavioral scoring considers the changes in the characteristics of consolidated clients and produces better results than those obtained with methodology used by the CNBV on provisions. The obtained results show the possibility of replacing the current model, minimizing the expected loss and increasing the ROA of Mexican financial institutions in 2.20% complying the methodology and statistical testing criteria according to the Unique Banking Dispositions and the guidelines of Basel II on credit risk.
Translated title of the contribution | Credit risk management at retail in Mexico: An econometric improvement in the selection of variables and changes in their characteristics |
---|---|
Original language | English |
Pages (from-to) | 377-398 |
Number of pages | 22 |
Journal | Contaduria y Administracion |
Volume | 62 |
Issue number | 2 |
DOIs | |
State | Published - Apr 2017 |
Keywords
- Banks
- Credit
- Data estimation methodology
- Econometric modeling
- Optimization techniques