• 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.96.2019.tde-27082019-140852
Document
Author
Full name
Renata Tavanielli
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Ribeirão Preto, 2019
Supervisor
Committee
Laurini, Marcio Poletti (President)
Caldeira, João Frois
Júnior, Sérgio Kannebley
Palma, Andreza Aparecida
Title in Portuguese
Modelo de taxa de juros a termo com mudanças de regime
Keywords in Portuguese
Estrutura a termo
MCMC
Mudanças de regime
Abstract in Portuguese
São propostas diversas extensões do modelo dinâmico de Nelson-Siegel para a análise do ajuste e da acurácia preditiva utilizando dados de DI do mercado brasileiro, incorporando, além de variáveis macroeconômicas, a possibilidade de mudanças de regime nos parâmetros dos modelos. As abordagens utilizadas são motivadas por evidências que sugerem mudanças de regime na curva de juros dos EUA, que é mais bem comportada que as de países em desenvolvimento, e evidências de mudança de regime em variáveis macroeconômicas brasileiras. Para a estimação verificamos que, além da incorporação de variáveis macroeconômicas, modelos com maior flexibilidade apresentam melhor ajuste. Na previsão fora da amostra, o desempenho dos modelos dependem do horizonte de previsão e da maturidade considerada, sendo que pelo procedimento do Model Confidence Set os modelos com mudança de regime se destacam. O modelo com mudança de regime nas variáveis latentes e nos fatores macroeconômicos (MDNS-MMacroEnd) destaca-se para maturidades mais elevadas para o horizonte de previsão de 1 mês. Para o horizonte de previsão de 12 meses, o modelo com mudança de regime baseado no artigo de So et al. (1998) nas variáveis macroeconômicas (MDNS-Smacro) apresenta melhor poder preditivo na maioria das maturidades analisadas. Já para o horizonte de 60 meses, o modelo com mudança de regime no fator de decaimento (MDNS-?) possui melhor acurácia preditiva para a maioria das maturidades
Title in English
Yield curve model with regime switches
Keywords in English
MCMC
Regime switch
Yield curve
Abstract in English
This thesis estimates versions of the dynamic Nelson-Siegel model, with macro factors and different methods of regime switch, for Brazilian ID data to analyze the fit and forecast accuracy. Our modeling approach is motivated by evidences suggesting breaks in the USA yield curve, which is more well behave than curves in developing countries, and evidences of regime switch in Brazilian macroeconomics variables. For the fit, the models with macro factors and more flexibility outperform the others. For out of sample forecast, the performance of the models depends of the forecast horizon and the maturity, in which by the procedure of the Model Confidence Set, models with regime switch stand out. Model with regime switch in the latent and macro factors (MDNS-MMacroEnd) outperforms in long maturities at shortest forecast horizon. For twelve month horizon, the model with regime switch based on So et al. (1998) in the macro factors (MDNS-Smacro) outperforms the others in most of the maturities. However, for the sixty month horizon, model with the regime switch in the decay factor (MDNS-?) has better predictive accuracy for most of the maturities
 
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
2019-10-22
 
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 © 2001-2024. All rights reserved.