Abstract

TREJO GARCIA, José Carlos; MARTINEZ GARCIA, Miguel Ángel  and  VENEGAS MARTINEZ, Francisco. Credit risk management at retail in Mexico: An econometric improvement in the selection of variables and changes in their characteristics. Contad. Adm [online]. 2017, vol.62, n.2, pp.377-398. ISSN 0186-1042.  https://doi.org/10.1016/j.cya.2017.01.003.

The early prediction of bad debtors for revolving credits in Mexico is a relevant issue today. The credit behavior econometric model proposed considers the changes in the characteristics of the consolidated accredited and provides better results than those obtained with the methodology utilized by the CNBV on provision matters. The results obtained show that the possibility of replacing the current model, minimizing the expected loss and increasing the ROA per financial institution at a national level by 2.20%, complies with the methodological criteria and the statistical tests in accordance with the Compiled Banking Regulation and Basel II guidelines on credit risk issues.

Keywords : Credit; Econometric modeling; Data estimation methodology; Optimization techniques.