Intraday volatility estimation in high-frequency data using order book information

• Autor(a) principal: Netze, Jules William Rudolf
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/70538
• topic: GARCH; Volatility estimation; High-frequency data; Limit order book; Domínio/Área Científica::Ciências Sociais::Economia e Gestão

This research conducts high-frequency intraday volatility estimations on the Euro Stoxx 50 Future under the multiplicative component GARCH framework, where the conditional volatility of high-frequency returns is decomposed into a daily, diurnal and stochastic intraday component. In contrast to existent research, this research covers a relatively long period of 423 trading days corresponding to about 345,000 1-minute observations. This study reveals that return series derived from the Limit Order Book have superior model features compared to simple trade returns. We find that these returns overcome the shortcomings of the welldocumented microstructure noise. Standardized residuals follow a white noise process and follow more closely a normal distribution compared to simple trade returns. However, this comes at the cost of larger coefficient instability and larger outliers in the estimated residuals.