A1056
Title: Stylized facts of cryptocurrency markets: Robust definitions and inference approaches
Authors: Nursultan Abdullaev - Innopolis University and Centre for Econometrics and Business Analytics SPbSU (Russia) [presenting]
Rustam Ibragimov - Imperial College Business School and Centre for Econometrics and Business Analytics SPbSU and New Economic School (United Kingdom)
Abstract: Several works in the literature have focused on the analysis of key stylised facts of cryptocurrency returns linked to fundamental problems of efficiency and predictability of cryptocurrency markets, including (i) heavy tails, indicating, in particular, that large price/return downfalls and fluctuations are more common than might be expected under a normal distribution; (ii) absence of autocorrelations, implying that return time series are to some extent are unpredictable and do not exhibit linear dependence over time; and (iii) volatility clustering, where periods of high volatility tend to be followed by similar periods and likewise for low volatility, implying nonlinear dependence in return time series. The contribution is the detailed study of the above properties of cryptocurrency markets using recently developed robust, econometrically and statistically justified definitions of and methods for inference on market (non-)efficiency, volatility clustering, and nonlinear dependence in return time series. In contrast to existing methodologies, the inference methods used in the analysis are robust against nonlinear dynamics and tail-heaviness of returns. The results of the analysis have significant implications for econometric modelling, risk management, and policy formulation in the context of cryptocurrency trading and investment.
