Expectile Hidden Markov Regression for Cryptocurrency Risk Analysis

Summary

The paper presents a linear expectile hidden Markov model for studying cryptocurrency returns in a risk management setting. Expectiles let the analysis focus on asymmetric parts of the return distribution, while model coefficients change over time according to an unobserved discrete Markov chain. The model uses an asymmetric normal distribution for estimation, and an Expectation-Maximization procedure with update formulas for its parameters.

The authors assess the approach using simulated data under several experimental settings and daily Bitcoin returns alongside major world market indices. The document describes the method and evaluation design, but does not report specific empirical findings or performance comparisons. Its scope is therefore methodological: the abstract does not establish that the model improves forecasts or risk decisions, nor does it specify how results vary across cryptocurrencies or market conditions.

Key ideas

• The model uses expectile regression to focus on asymmetric features of cryptocurrency returns.
• Latent Markov states allow regression coefficients to evolve over time.
• An asymmetric normal distribution supports maximum likelihood estimation.
• An Expectation-Maximization algorithm supplies parameter updates.
• The evaluation covers simulated data and Bitcoin returns in relation to major market indices.

Tags

# Expectile hidden Markov regression models for analyzing cryptocurrency returns


# Expectile hidden Markov regression models for analyzing cryptocurrency returns









In this paper we develop a linear expectile hidden Markov model for the analysis of cryptocurrency time series in a risk management framework. The methodology proposed allows to focus on extreme returns and describe their temporal evolution by introducing in the model time-dependent coefficients evolving according to a latent discrete homogeneous Markov chain. As it is often used in the expectile literature, estimation of the model parameters is based on the asymmetric normal distribution. Maximum likelihood estimates are obtained via an Expectation-Maximization algorithm using efficient M-step update formulas for all parameters. We evaluate the introduced method with both artificial data under several experimental settings and real data investigating the relationship between daily Bitcoin returns and major world market indices.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.