Title: Price endogeneity of the cryptocurrency market
Authors: Jan Sila - UTIA AV CR, v.v.i. (Czech Republic) [presenting]
Michael Mark - Ecole Polytechnique Federale Lausanne (Switzerland)
Abstract: A novel source of high-frequency data is explored by the means of temporal point processes. We propose a non-parametric mutually exciting Hawkes process to examine the inner and cross dependencies of transactions for four major cryptocurrencies on the BitMEX exchange. As these instruments are considered being highly volatile and sensitive to news releases, we enrich the model with a kernel matrix representing the news impact. Thus introducing an original extension of the Hawkes process by combining news and mutual excitations of the series. This allows for a realistic time-dependent measure of the degree of endogeneity in the market.