CMStatistics 2023: Start Registration
View Submission - CMStatistics
B0688
Title: Bivariate CARMA-Hawkes model: Theory and applications Authors:  Edit Rroji - Universita' degli studi di Milano-Bicocca (Italy) [presenting]
Lorenzo Mercuri - University of Milan (Italy)
Andrea Perchiazzo - Vrije Universiteit Brussel (Belgium)
Abstract: The differences between bonds labelled as green and brown from the lens of the trading activity are investigated, especially the idea that positive and negative jumps in the dynamics of returns have a specific memory nature that can be modelled through a self-exciting process. Specifically, the properties of high-frequency time series of brown and green bonds are studied using Hawkes processes where the kernel is a CARMA(p,q) model. Besides, findings point out that the intensities of positive and negative jumps in the price dynamics are not stationary through time and that better fitting results in our data set, especially for the issuer that operates in the energy market, are achieved by means of a higher order of bivariate Hawkes models.