Volatility and spillover effects in stock markets: A multivariate GARCH approach

• Autor(a) principal: Marques, Bernardo Rodrigues
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/27025
• topic: Market indices; Volatilidade -- Volatility; Spillover effects; GARCH models; Multivariate GARCH; Índices de mercado; Efeitos de contágio; Modelos GARCH; GARCH multivariado

The goal of this dissertation is to study the behavior of financial market’s volatility. Performing univariate and multivariate analyses, we were able to find the models that can more accurately predict future market indices’ volatilities, considering the spillover effects between them. The data used in this study are the daily prices of S&P 500, DAX and Nikkei 225 indices, that replicate the North American, German and Japanese markets, respectively. The practical part can be divided into two phases. The first part is a univariate analysis of the indices, in which we make the initial diagnosis of the data, specify the conditional mean equation, test the existence of conditional heteroscedasticity in the process and estimate the univariate GARCH models. In a second part, more dedicated to the joint analysis of the indices, we start by applying the pairwise Granger causality test, estimate and diagnose the VAR models, estimate the multivariate GARCH models and take conclusions. Throughout the process, we use information criteria to make decisions. The results indicate that the volatility of the indices tends to be higher in periods of economic crisis and lower in periods of stability. When the goal is to predict the future behavior of the volatility of these market indices separately, we should use the EGARCH(1,1) model. Between the indices under analysis, S&P 500 and DAX are the pair with more presence of spillover effects between them. If we want to forecast market volatilities considering spillover effects, the most suitable model is the symmetric DCC model.