Reasoning and Purpose
The existence of market opportunities around sudden price moves required the need for speed in market participants’ reactions, which is technologically enforced via High Frequency Trading (HFT). This increased interest in the understanding of fast dynamics, as well as the availability of high-frequency data generated an academic interest in the identification of the general properties of market moves at high and ultra high frequencies. If until the end of the 90’s the focus was mostly on the daily frequency (academic reference on the stylized facts of daily stock market returns is Cont 2001, Mantegna & Stanley 2000, Dacorogna et. al. 2001), the 2000’s brought a wide selection of research papers analyzing the dynamics of intra-day returns at high and ultra-high frequencies, lately going to the level of tick-by-tick data.
Remarkable events represent a field of research that enjoyed important attention lately by the study of jump detection techniques for both univariate and multivariate series. The numerous tests that were generated by this stream of investigations is fed by the interest to detect dynamics that can be considered out-of-ordinary.
We share the view that the market has the tendency to test for existence of such events, to identify their occurence and initiate trading strategies based on their local or general properties.
Our objective is to provide a framework for the implementation of the methodologies existing in the current academic literature, identify their practical usefulness and reflect on their possible automatization.