Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Estudo Geralarrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Estudo Geral
Master thesis . 2022
Data sources: Estudo Geral
addClaim

Heteroscedasticidade Condicional Auto-Regressiva Generalizada na modelação de séries temporais

Authors: Monteiro, Inês Marques;

Heteroscedasticidade Condicional Auto-Regressiva Generalizada na modelação de séries temporais

Abstract

Os processos estocásticos com formulações Auto-Regressivas Condicionalmente Heteroscedásticas(ARCH), introduzidos por Engle em 1982, têm vindo a desempenhar um papel importante na análisede séries temporais. O estudo destes modelos além de ter sido muito útil para diversas áreas científicas tem vindo a ter especial interesse no domínio das séries de natureza financeira.Em 1986, Bollerslev propõe uma generalização dos modelos ARCH, que designou por GARCH. Desde então têm surgido na literatura muitas variantes destes modelos.Nesse sentido, o presente trabalho inclui uma breve descrição de algumas das principais variações e extensões dos modelos GARCH, mais concretamente os modelos EGARCH (modelos GARCHexponenciais) e os modelos GTARCH (modelos GARCH com níveis).Será ainda apresentado o modelo GARCH (1,1), base do estudo dos modelos GARCH clássicos,e uma nova variante de tais modelos, recentemente proposta por Voutilainen et al. (2021). Seguindo o estudo destes autores, serão obtidas condições que assegurem a existência e a unicidade de soluções estritamente estacionárias e fracamente estacionárias. Incluído nesse estudo está também a construção de estimadores para os parâmetros do modelo com base em funções de autocovariância. Para tal, considera-se uma caracterização AR(1) do modelo que conduz a equações quadráticas do tipo Yule-Walker. Provar-se-á que esta abordagem permite obter estimadores com boas propriedades. A fim de avaliar o comportamento dos estimadores obtidospara os parâmetros do modelo serão efetuados estudos de simulação.Para terminar, consideramos um conjunto de dados reais correspondentes à empresa Corticeira Amorim. Procedemos ao ajustamento dos dados através de um modelo da classe dos GARCH clássicos, recorrendo ao software estatístico Eviews, e através desta nova proposta de Voutilainen et al.(2021), recorrendo ao MATLAB. Os modelos estimados resultantes serão brevemente comparados.

Stochastic processes with Autoregressive Conditionally Heteroscedastic (ARCH) formulations, introduced by Engle in 1982, have been playing an important role in time series analysis. The study of these models, in addition to having been very useful for several scientific areas, has been of special interest in the field of financial series.In 1986, Bollerslev proposed a generalization of the ARCH models, which he designated byGARCH. Since then, many variants of these models have appeared in the literature.In this sense, this paper includes a brief description of some of the main variations and extensions of the GARCH models, more specifically the EGARCH models (exponential GARCH models) and the GTARCH models (GARCH models with levels).We will also present the GARCH (1,1) model, the basis of the study of classical GARCH models, and a new variant of such models, recently proposed by Voutilainen et al. (2021). Following thestudy of these authors, we will obtain conditions that ensure the existence and uniqueness of strictly stationary and weakly stationary solutions. Included in this study is also the construction of estimators for the model parameters based on autocovariance functions. For this, an AR(1) characterization of the model that leads to quadratic equations of the Yule-Walker type is considered. It will be proved that this approach leads to estimators with good properties. In order to evaluate the behavior of the estimators obtained for the model parameters, simulation studies will be carried out.Finally, we consider a set of real data corresponding to the company Corticeira Amorim. Weproceed to the adjustment of the data through a model of the classic GARCH class, using the statistical software Eviews, and through this new proposal by Voutilainen et al. (2021), using MATLAB. The resulting estimated models are briefly compared.

Dissertação de Mestrado em Matemática apresentada à Faculdade de Ciências e Tecnologia

Country
Portugal
Related Organizations
Keywords

Séries Temporais, Conditionally Heteroscedastic Autoregressive Formulations, Estacionaridade Fraca, Weak Stationarity, Estimators, Formulações Auto-Regressivas Condicionalmente Heteroscedásticas, Estacionaridade Estrita, Time Series, Strict Stationarity, Estimadores

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Green