4th International Conference on Econometrics and Statistics, Kowloon, Hong Kong, 22 - 26 June 2021, pp.1-19
The self-excitability and price clustering properties of the cryptocurrency market are studied to investigate the main sources of volatility, in particular, the reflexivity or the endogeneity issues. We apply our kernel estimation of the spectrum localized both in time and frequency to data sets of transaction times, revealing pertinent features in the data that had not been made visible by classical non-localized approaches based on models with constant fertility functions over time. We apply the empirical analysis to the three largest crypto assets, i.e. Bitcoin - Ethereum - Ripple, and provide a comparison with other financial assets such as SP500, Gold, and the volatility index VIX observed from January 2018 to December 2020. The results show high levels of endogeneity in the basket of cryptocurrencies under investigation, underlining the evidence of a significant role of endogenous feedback mechanisms in the price formation process. We also demonstrate that the level of the endogeneity of markets, quantified by the branching ratio of the Hawkes process, is overestimated if the time variation is not considered.