发表机构
Inha University; Korea Institute for Advanced Study（仁荷大学; 韩国高等研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过CECP和directed HVG分析2020-2025年高频数据，发现加密货币与股票市场典型事实表面收敛但动力学过程根本不同，为两类市场的差异研究提供了新证据。
AI 中文摘要
本研究探究加密货币的宏观统计成熟度是否意味着其与传统股票市场具有动力学等价性。我们使用复杂度-熵因果平面（CECP）和有向水平可见性图（directed HVG）分析2020至2025年的高频数据，以揭示收益率序列中的复杂时间模式和时间导向结构。尽管常规典型事实在所有资产间呈现显著收敛性，但结构诊断揭示了一个引人注目的悖论：加密货币在普通时期比股票基准更具局部随机性，然而在高可见性收益率事件周围表现出明显更强的时间不可逆性。绝对收益率结果显示，加密货币的大幅波动往往突然开始并在之后保持高位。对正收益率和负收益率幅度的单独分析表明，该模式在加密货币的上行方向是一致的，但在下行方向因资产而异。我们得出结论：统计成熟度仅为表面现象；成熟加密货币的潜在动力学过程与传统基准仍存在根本差异。
英文摘要
This study investigates whether the macroscopic statistical maturity of cryptocurrencies implies dynamical equivalence with an equity-market benchmark. We analyze one-minute data (2020--2025) for Bitcoin, Ethereum, XRP, and E-mini S\&P~500 futures using the Complexity--Entropy Causality Plane (CECP) and directed horizontal visibility graphs (DHVG). While conventional stylized facts overlap across the analyzed assets, cryptocurrencies exhibit weaker local ordinal organization and a larger divergence between their high-degree in- and out-visibility distributions relative to matched surrogate baselines under the primary specification. This high-degree separation is strongest at the one-minute scale and weakens under 5- and 10-minute aggregation, while the direction of the underlying asymmetry remains more asset- and preprocessing-dependent. We therefore conclude only that macroscopic similarity can coexist with diagnostic-specific structural differences in the assets and period examined.
