IDENTIFICAÇÃO DE MODELOS GARCH USANDO INTELIGÊNCIA COMPUTACIONAL / GARCH MODELS IDENTIFICATION USING COMPUTATIONAL INTELLIGENCE

AUTOR(ES)

ANDRE MACHADO CALDEIRA

DATA DE PUBLICAÇÃO

2009

RESUMO

ARCH and GARCH models have been largely explored technically and empirically since their creation in 1982 and 1986, respectively. However, the focus has always been on stylized facts of financial time series or volatility forecasts, where GARCH(1,1) has commonly been used. Studies on identification of GARCH models have been rare. In this context, this work aims to develop an intelligent system for improving the specification of GARCH models, thus avoiding the indiscriminate use of the GARCH(1,1) model. In order to validate the efficacy of the proposed system, simulated time series are used. Results are compared to chosen models through AIC and BIC criteria. Their performances are then compared by using real data.

ASSUNTO(S)

redes neurais identificacao computational intelligence inteligencia computacional neural networks identification

Documentos Relacionados

• Structural damage identification via accoustic wave propagation approach using computational intelligence techniques
• Identificação de locutor usando modelos de misturas de gaussianas.
• Estimação e previsão da estrutura a termo das taxas de juros usando técnicas de inteligência computacional
• INTELIGÊNCIA COMPUTACIONAL APLICADA EM MACHINE LEARNING
• Estimação indireta de modelos R-GARCH