Study of volatility structures in geophysics and finance using GARCH models
This work investigates the underlying volatility processes in earthquake series, high frequency (tick) data, financial indice and explosive series. Furthermore it examines the applicability of a range of GARCH specifications for modeling volatility of these series in order to identify similarities and differences in the volatility structures. The GARCH variants considered include the basic GARCH, IGARCH, ARFIMA (0,d,0)-GARCH and FIGARCH specifications. The methodology is not new, however the major contribution of this work comes in the realm of applications. The methodology is applied to three domains: Geophysics (earthquake data), Finance (high frequency data and indices) and explosives data. In all the applications the methodology provides insight into features of these series volatility. ^ The results show that the FIGARCH specification is favoured in the DJIA, S&-P500 and the explosives series volatility but in the BAC and JPM high-frequency data and in the earthquake series the ARFIMA-GARCH specification is preferred as it more reliably describes the volatility of these series. The WMT and IBM high-frequency data volatility were best described using the GARCH model. ^
Biney, Francis, "Study of volatility structures in geophysics and finance using GARCH models" (2012). ETD Collection for University of Texas, El Paso. AAI1512549.