CMStatistics 2018: Start Registration
View Submission - CFE
Title: A multivariate dynamic mixture model for discrete price changes at high frequency Authors:  Paolo Santucci de Magistris - LUISS Guido Carli (Italy)
Leopoldo Catania - Aarhus BBS (Denmark) [presenting]
Roberto Di Mari - University of Rome Tor Vergata (Italy)
Abstract: High frequency price changes of financial assets are usually assumed to follow a distribution with continuous support and time-varying parameters. However, the tick structure of the financial markets entails that price changes observed at very high frequency are discrete. We start from this empirical evidence to develop a new model able to describe the dynamic properties of a multivariate time-series of high frequency price changes, including the high probability of observing no variations (zeroes). We assume the existence of two independent latent processes determining the dynamic properties of the price changes and the probability of the occurrence of zeroes. Given the probabilistic structure embedded in our modelling framework we analyze the different sources of this large amount of zeroes as for example: absence of news, same magnitude of positive and negative news, and periods of market illiquidity. Furthermore, we propose a multivariate model to investigate the dynamics of the zeroes across several assets.