Assessing systemic importance with a fuzzy logic inference system

• Autores: Carlos León, Clara Machado, Andrés Murcia
• Localización: Intelligent systems in accounting, finance & management, ISSN-e 1099-1174, Vol. 23, Nº. 1-2, 2016, págs. 121-153
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Three metrics are designed to assess Colombian financial institutions' size, connectedness and non-substitutability as the main drivers of systemic importance: (i) centrality as net borrower in the money market network; (ii) centrality as payments originator in the large-value payment system network; and (iii) asset value of core financial services. An aggregated systemic importance index is calculated based on expert knowledge by using a fuzzy logic inference system. We use principal component analysis to calculate a benchmark index for comparison purposes. Overall similarities between both indexes put forward that expert knowledge aggregation is consistent with that based on a purely quantitative standard approach. Specific non-negligible differences concur with the nonlinear features of an approach whose intention is to replicate human reasoning. Both indexes are complementary and provide a comprehensive relative assessment of each financial institution's systemic importance in the Colombian case, in which the choice of metrics pursues the macroprudential perspective of financial stability.