A Predictive Framework Integrating Multi-Scale Volatility Components and Time-Varying Quantile Spillovers: Evidence from the Cryptocurrency Market

TL;DR

This paper introduces a new volatility forecasting framework for cryptocurrencies that captures asymmetric spillovers and decoupling effects, improving prediction accuracy especially during extreme market conditions.

Contribution

It develops the SA-Log-HAR model integrating time-varying quantile spillovers and demonstrates its superior forecasting performance over existing models.

Findings

Uncovered asymmetric volatility spillovers in cryptocurrency markets.

Identified structural decoupling between market size and systemic importance.

SA-Log-HAR outperforms benchmarks in extreme volatility prediction.

Abstract

This paper investigates the dynamics of risk transmission in cryptocurrency markets and proposes a novel framework for volatility forecasting. The framework uncovers two key empirical facts: the asymmetric amplification of volatility spillovers in both tails, and a structural decoupling between market size and systemic importance. Building on these insights, we develop a state-adaptive volatility forecasting model by extracting time-varying quantile spillover features across different volatility components. These features are embedded into an extended Log-HAR structure, resulting in the SA-Log-HAR model. Empirical results demonstrate that the proposed model outperforms benchmark alternatives in both in-sample fitting and out-of-sample forecasting, particularly in capturing extreme volatility and tail risks with greater robustness and explanatory power.