Time Series Modeling of Earthquake Ground Motions Using ARMA-GARCH Models

Abstract

Engineers are well aware that, due to the stochastic nature of earthquake ground motion, the information obtained from structural response analysis using scant records is quite unreliable. Thus, providing earthquake models for specific sites or areas of research and practical implementation is essential. This paper presents a procedure for the modeling strong earthquake ground motion based on autoregressive moving average (ARMA) models. The Generalized autoregressive conditional heteroskedasticity (GARCH) model is used to simulate the time-varying characteristics of earthquakes.