Estimar el capital económico basado en riesgos con los métodos de agregación de matrices de correlación y cópulas
Loading...
Código QR
Authors
Rau Campos, Selena Ana
Abstract
En 2019, la Superintendencia de Banca, Seguros y AFP (SBS) propuso un nuevo modelo de solvencia, específicamente de capital basado en riesgos, con el objetivo de mantener una adecuada solvencia en las empresas de seguros, garantizando así que cuenten con los recursos suficientes para afrontar escenarios extremos de pérdidas. Para el cálculo de los capitales, la SBS, en calidad de ente regulador, planteó una metodología de agregación mediante matriz de correlación; sin embargo, las compañías de seguros deberían analizar otras alternativas de agregación para el cálculo de capital. Una metodología ampliamente utilizada para este fin es la agregación por cópulas, la cual se aplicó en este análisis (específicamente, la t-cópula). En ambas metodologías, se necesitó estimar las reservas matemáticas best estimate, incluidas en el pasivo de los estados financieros de la compañía. El beneficio de diversificación al utilizar la metodología de agregación de matrices de correlación fue de aproximadamente USD 516 000, mientras que para la agregación por t-cópulas fue de aproximadamente USD 532 000. Con estos resultados, se puede concluir que, a pesar de haber utilizado los mismos parámetros, la agregación por t-cópula permite obtener un mayor beneficio de diversificación para la empresa, liberando capital y permitiendo su empleo en inversiones con una rentabilidad esperada más alta.
In 2019, the Superintendency of Banking, Insurance and AFP (SBS) proposed a new solvency model, specifically risk-based capital, with the objective of maintaining adequate solvency in insurance companies, thus ensuring that they have sufficient resources to face extreme loss scenarios. For the calculation of capital, the SBS, in its capacity as regulator, proposed an aggregation methodology using a correlation matrix; however, insurance companies should analyze other aggregation alternatives for the calculation of capital. A widely used methodology for this purpose is the copula aggregation, which was applied in this analysis (specifically, the t-copula). In both methodologies, it was necessary to estimate the best estimate mathematical reserves, included in the liabilities of the company's financial statements. The diversification benefit of using the correlation matrix aggregation methodology was approximately USD 516 000, while for t-copula aggregation it was approximately USD 532 000. With these results, it can be concluded that, despite having used the same parameters, the t -copula aggregation allows obtaining a greater diversification benefit for the company, freeing up capital and allowing its use in investments with a higher expected return.
In 2019, the Superintendency of Banking, Insurance and AFP (SBS) proposed a new solvency model, specifically risk-based capital, with the objective of maintaining adequate solvency in insurance companies, thus ensuring that they have sufficient resources to face extreme loss scenarios. For the calculation of capital, the SBS, in its capacity as regulator, proposed an aggregation methodology using a correlation matrix; however, insurance companies should analyze other aggregation alternatives for the calculation of capital. A widely used methodology for this purpose is the copula aggregation, which was applied in this analysis (specifically, the t-copula). In both methodologies, it was necessary to estimate the best estimate mathematical reserves, included in the liabilities of the company's financial statements. The diversification benefit of using the correlation matrix aggregation methodology was approximately USD 516 000, while for t-copula aggregation it was approximately USD 532 000. With these results, it can be concluded that, despite having used the same parameters, the t -copula aggregation allows obtaining a greater diversification benefit for the company, freeing up capital and allowing its use in investments with a higher expected return.
Description
Universidad Nacional Agraria La Molina. Facultad de Economía y Planificación. Departamento Académico de Estadística e Informática
Keywords
Mejor estimación; matriz de correlación
Citation
Date
2024
Collections
Seleccionar año de consulta:
Licencia de uso
Excepto si se señala otra cosa, la licencia del ítem se describe como info:eu-repo/semantics/openAccess