• 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.92.2005.tde-15092022-095633
Document
Author
Full name
André Beraha
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2005
Supervisor
Committee
Francisco, Gerson (President)
Ferreira, Fernando Fagundes
Vicente, Renato
Title in Portuguese
Aplicação de redes neurais em risco operacional: análise discriminante em perdas trabalhistas
Keywords in Portuguese
Análise de risco
Análise discriminante
Redes neurais
Abstract in Portuguese
Com a publicação do novo acordo de capital da Basiléia, as instituições financeiras incluíram no seu processo de gerenciamento e mensuração de riscos um novo componente importante : o risco operacional. As instituições financeiras têm despendido nos últimos anos um esforço considerável na modelagem do risco operacional, tanto em termos de cálculo de capital quanto em modelos quantitativos de controle e gestão desse tipo de risco específico. Neste estudo, será implementada a técnica de redes neurais sobre uma importante categoria de evento de risco operacional denominada práticas empregatícias, que tem como efeito as perdas trabalhistas. A análise discriminante baseada em redes neurais a ser tratada neste estudo tem o intuito de modelar a probabilidade que um indivíduo possui de gerar perdas trabalhistas, dado um conjunto de características dele. Além disso, o modelo deve apresentar um boa performance de classificação desse indivíduo nas duas classes disponíveis (gerar ou não perda trabalhista). Os resultados deste estudo motivam futuras pesquisas a utilizarem modelos quantitativos para controle, mensuração e gerenciamento de risco operacional.
Title in English
Application of neural networks in operational risk discriminant analysis in labor losses
Keywords in English
Discriminant analysis
Neural networks
Risk analysis
Abstract in English
With the publishing of the new basel capital accord, the financial institutions have included an important new component in their process of management and measurement of risks: operational risk. In the last few years, financial institutions have devoted tremendous efforts towards modeling operational risk both in terms of calculation of capital and of quantitative models of control and management regarding this type of specific risk. In this study the technique of artificial neural networks will be applied to an important category of operational risk event called employment practices, whose one of effects are labor losses. The discriminating analysis based on artificial neural networks addressed in this study is designed to model the probability an individual has of generating labor losses given a set of individual features. In addition, the model should have a good classification performance of the individuals in to the two classes available (generating or not generating labor losses). The results of this study encourage future research in using quantitative models to control, measure and manage operational risk.
 
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.
MpAndreBeraha.pdf (1.45 Mbytes)
Publishing Date
2022-09-15
 
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 © 2001-2024. All rights reserved.