AI 中文总结
研究金融市场中机构声明、媒体回声与交易定位的关系，提出线性二次模型和测量方法，证明负Say-Do协方差时言辞应被淡化，并在模拟市场中验证了回声情绪的负预测性和工具的有效性。
AI 中文摘要
基于金融文本构建的机器学习信号将机构所言及媒体所复述视为价值的证据。但塑造叙事者可能正与其交易方向相反。我们研究具有三种可观察声音的市场：机构声明（Say）、媒体回声（Echo）和揭示性定位（Do）。我们探究何时应追随言辞，何时应淡化言辞。在一个线性二次模型中，知情的机构在部分轻信的群体面前发言并交易，当且仅当 $\varphi^2<2\lambda k<\varphi$ 时，压低资产言论同时买入该资产为最优策略。随后，一个无分布恒等式表明，当可观察的 Say-Do 协方差为负时，言辞具有负向预测内容，应予以淡化。在测量方面，我们推导出（i）一个精确的因子化后验，用于识别哪些文章是回声，结合到达时间与嵌入相似度；（ii）一个收益对齐的对比目标，当平方嵌入距离是平方结果距离的递增仿射函数时，该目标恰好达到其界限，并提供最紧凑的基于损失的证书，以确定哪些邻居排序在不完美训练中幸存；（iii）一个路径签名统计量，用于判断谁先行动。在具有已知真实标签的受控市场中，回声情绪在所有29个模拟市场中以显著负号预测收益，滚动 Say-Do 相关性以0.90的AUC标记误报事件，收益对齐的嵌入按后果而非主题组织头条新闻。我们还报告了这些工具的失效场景。
英文摘要
Machine-learning signals built from financial text treat what institutions say, and what the media repeat, as evidence about value. But whoever shapes a narrative may be trading against it. We study markets with three observable voices: institutional statements (Say), media repetition (Echo) and revealed positioning (Do). We ask when words should be followed and when they should be faded. In a linear-quadratic model of an informed institution that speaks and trades before a partly credulous crowd, talking an asset down while buying it is optimal exactly when $φ^2<2λk<φ$. A distribution-free identity then shows that when the observable Say-Do covariance is negative, words carry negative predictive content and should be faded. For measurement, we derive (i) an exact factorised posterior over which articles are echoes, combining arrival times with embedding similarity; (ii) a return-aligned contrastive objective that attains its bound exactly when squared embedding distances are an increasing affine function of squared outcome distances, with the tightest loss-based certificate of which neighbour rankings survive imperfect training; and (iii) a path-signature statistic for who moved first. In a controlled market with known ground truth, echo sentiment predicts returns with a significantly negative sign in all 29 simulated markets, the rolling Say-Do correlation flags false-alarm events with an AUC of 0.90, and return-aligned embeddings organise headlines by consequence rather than topic. We also report where the tools fail.
CommentsFirst draft. 9 pages main text, 31 pages total, 3 figures in the main text. Code: https://github.com/AliAtiah/say-echo-do
