• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.92.2002.tde-22122021-120057
Document
Author
Full name
Artur Henrique de Toledo Damasceno
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2002
Supervisor
Committee
Pereira, Carlos Alberto de Braganca (President)
Dreifus, Henrique Von
Securato, Jose Roberto
Title in Portuguese
Modelo de cálculo de risco bancário: caso da crise financeira equatoriana de 1999/2000
Keywords in Portuguese
Econometria
Estatistica aplicada
Inferência Bayesiana
Abstract in Portuguese
Com a perspectiva semelhante a de um órgão de controle (Banco Central ou Superintendência de Bancos), este trabalho propõem um modelo para monitoramento periódico de cada banco do setor financeiro, de forma comparativa. Os dados dos bancos do sistema financeiro equatoriano foram analisados neste trabalho, buscando diferenças estruturais e estratégicas "ex-ante" sobre aqueles que vieram a sofrer intervenções a partir de abril de 1999, como fruto dos fortes choques externos sofridos pela economia daquele país no período 1997 - 1998. Como variável resposta do modelo proposto, têm-se as próprias intervenções realizadas. Pelo lado das variáveis preditoras, apresentavam-se cerca de 48 índices financeiros e contábeis mensalmente calculáveis para cada banco. Dentre estes índices, analisados para o período setembro de 1996 a março de 1999, apenas 6 foram efetivamente considerados na modelagem, por diferenciarem as distribuições dos bancos em curso normal daquelas dos bancos com problemas. Estes seriam indicativos de que os dois tipos de bancos já operavam de forma diferenciada bem antes das intervenções. As séries das variáveis preditoras selecionadas foram reduzidas através das estatísticas regressão linear e média de 24 meses. A partir disso, puderam-se modelar e calcular dois Índices Principais, aqui denominados por IPR e lPM, com maior poder de predição do que pelas variáveis isoladamente. Quando comparam-se estes Índices Principais à lista de bancos que sofreram intervenção, o resultado é bastante satisfatório, com uma leve vantagem para lPM. Baseando-se nos dados amostrais e suas medidas, e no cálculo do índice lPM, utilizou-se o Método Bayesiano para cálculo da probabilidade, ou risco de intervenção.
Title in English
Bank risk calculation model: case of the Ecuadorian financial crisis of 1999/2000
Keywords in English
Applied Statistics
Bayesian Inference
Econometrics
Abstract in English
With the perspective similar to the one from a control and regulatory agency (Central Bank or Supervision of Banks), this paper proposes a model to periodically monitor each bank from the financial sector in a comparative way. Information related to the banks from the Ecuadorian financial system were analysed, in a search for "ex-ante" structural and strategic differences among the banks that suffered interventions as from April 1999 on, as a consequence of the strong external shocks suffered by the economy of that country in period 1997 1998. Interventions themselves were considered as output variable for the proposed model. Concerning the input variables, there were about 48 financial and accounting ratios, all possible to be monthly calculated per bank. Amongst them, analysed for the period September of 1996 to March of 1999, only 6 could be effectively considered in this modelling, for differentiating the distributions of the banks in normal course from those of the banks with problems. Such results could indicate that these two types of banks were already operating differently before the interventions. The series of the selected input variables were reduced through the statistics of a linear regression and of an average of 24 months. After that, two Indexes (Índice Principal) were modelled and calculated, the lPR and the IPM, with a better level of prediction than of any of the input variables separately. When these indexes were compared with the list of banks that had suffered intervention, the results were quite satisfactory, with a slight advantage for lPM. Based on the sample data and its measures, and with the calculation of lPM the Bayesian Method was applied to compute the probability, the risk of intervention.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2021-12-22
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2022. All rights reserved.