• 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.3.2016.tde-14072016-152635
Document
Author
Full name
Marcelo Rosario da Barrosa
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2015
Supervisor
Committee
Ribeiro, Celma de Oliveira (President)
Chela, João Luiz
Tancredi, Thiago Pontin
Title in Portuguese
Aplicação de métodos computacionais multidisciplinares de engenharia para otimização de carteiras de investimentos.
Keywords in Portuguese
Estatísticas de ordem
Gestão de carteiras
Interferência não paramétrica
Interpolação estatística
Investimentos
Otimização global
Redes Neurais
Risco
Abstract in Portuguese
Este trabalho apresenta uma nova metodologia para otimizar carteiras de ativos financeiros. A metodologia proposta, baseada em interpoladores universais tais quais as Redes Neurais Artificiais e a Krigagem, permite aproximar a superfície de risco e consequentemente a solução do problema de otimização associado a ela de forma generalizada e aplicável a qualquer medida de risco disponível na literatura. Além disto, a metodologia sugerida permite que sejam relaxadas hipóteses restritivas inerentes às metodologias existentes, simplificando o problema de otimização e permitindo que sejam estimados os erros na aproximação da superfície de risco. Ilustrativamente, aplica-se a metodologia proposta ao problema de composição de carteiras com a Variância (controle), o Valor-em-Risco (VaR) e o Valor-em-Risco Condicional (CVaR) como funções objetivo. Os resultados são comparados àqueles obtidos pelos modelos de Markowitz e Rockafellar, respectivamente.
Title in English
Application of multidisciplinary engineering methods to optimize investment portfolios.
Keywords in English
Kriging
Neural networks
Optimization
Portfolio management
Risk
Abstract in English
This work presents a new methodology for optimizing financial asset portfolios. The proposed methodology, based on universal interpolators such as Artificial Neural Networks and the Kriging Method, allows for approximating the risk surface - and thus the optimal solution to the problem - in a generalized fashion and applicable to any risk measure known in literature, relaxing every restrictive hypothesis inherent to the available methods and with the ability to estimate the error in the approximation. Illustratively, the proposed methodology is applied to the portfolio problem with the Variance (control), Value-at-Risk and Conditional Value-at-Risk as objective functions. Results are compared to those obtained by Markowitz and Rockafellar models, respectively.
 
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
2016-07-15
 
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.