Detection of low-frequency large-amplitude jump in financial time series

The continuous-time and discrete-time generalized market microstructure (GMMS) model are proposed for describing the dynamics of non-Gaussian financial time series. The GMMS model is a class of jump-diffusion model that can represent the dynamic behaviors of measurable market price, immeasurable market excess demand and market liquidity, and also the relationship of three variates in a market. The model includes a jump component that is used to capture the large abnormal variations of financial assets, which could occur when market is affected by some special events happened suddenly, such as release of important financial information. On the basis of the discrete-time GMMS model, a detection algorithm of low-frequency and large-amplitude jump component is presented, which is developed in accordance with the Markov property of financial time series and the Bayes´ theorem. Both simulation and case study verify the effectiveness of the model and its estimation approach proposed in this paper.