Stochastic volatility models describe asset prices [Formula: see text] as driven by an unobserved process capturing the random dynamics of volatility [Formula: see text]. We quantify how much information about [Formula: see text] can be inferred from asset prices [Formula: see text] in terms of Shannon’s mutual information in a twofold way: theoretically, by means of a thorough study of Heston’s model; from a machine learning perspective, by means of investigating a family of exponential Ornstein–Uhlenbeck (OU) processes fitted on S&P 500 data.

中文翻译：

---

随机波动率模型的信息理论分析

随机波动率模型将资产价格 [公式：参见文本] 描述为由捕获波动率随机动态的未观察到的过程驱动 [公式：参见文本]。我们从香农互信息的角度，从资产价格[公式：见文本]中推断出多少关于[公式：见文本]的信息，从两个方面进行量化：理论上，通过对赫斯顿模型的深入研究；从机器学习的角度来看，通过调查适合标准普尔 500 指数数据的一系列指数 Ornstein-Uhlenbeck (OU) 过程。