• 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.12.2004.tde-07122021-095621
Document
Author
Full name
Mauri Aparecido de Oliveira
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2004
Supervisor
Committee
Siqueira, Jose de Oliveira (President)
Canton, Adolpho Walter Pimazoni
Rezende, Solange Oliveira
Title in Portuguese
Previsão de sucessões cronológicas econômico-financeiras por meio de redes neurais artificiais recorrentes de tempo real e de processos ARMA-GARCH: um estudo comparativo quanto à eficiência de previsão
Keywords in Portuguese
Análise de séries temporais
Econometria
Previsão (análise de séries temporais)
Redes neurais
Abstract in Portuguese
O principal objetivo desse trabalho é estudar o processamento de séries temporais para a realização de previsão utilizando redes neurais artificiais e os modelos ARIMA-GARCH. Com relação as redes neurais foram estudados os algoritmos de processamento temporal utilizando redes neurais alimentadas adiante e as redes recorrentes. Sendo que nas redes recorrentes o algoritmo utilizado para análise da série temporal foi o algoritmo de aprendizagem recorrente em tempo real (RTRL). Para os modelos ARIMA foi utilizada a metodologia desenvolvida por Box e Jenkins. Foram utilizadas as séries temporais de retornos diários do IBOVESPA, Petrobrás, Nasdaq, IBM e saca de 60Kg de soja como exemplo de aplicação das metodologias
Title in English
Forecasting economic-financial chronological successions using real-time recurrent artificial neural networks and ARMA-GARCH processes: a comparative efficiency study
Keywords in English
Econometrics
Neural networks
Prediction (time series analysis)
Time series analysis
Abstract in English
The main objective of this dissertation is the time series processing to perform forecasting using artificial neural networks and ARIMA models. Regarding to neural networks to perform time series processing my studies focused feedforward and recurrent networks. The main recurrent algorithm applied to time series analysis were real time recurrent learning (RTRL). The Box and Jenkins methodology was applied to ARIMA analysis. As an application example we analyzed the following daily returns time series: IBOVESPA, Petrobras, Nasdaq, IBM, 60Kg soybean bag
 
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
2021-12-08
 
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.