Market Impact Estimation via Causal Foundation Models
作者: Dennis Thumm, Ying Chen
分类: cs.LG, cs.CE, q-fin.TR
发布日期: 2026-10-08
备注: In 7th ACM International Conference on AI in Finance (ICAIF '26), November 14-17, 2026, Milan, Italy
💡 一句话要点
提出因果市场模拟器以解决市场影响估计问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 因果建模 市场影响 金融交易 机器学习 数据拟合
📋 核心要点
- 现有的市场影响估计方法在处理信息流对价格的影响时存在偏差，无法准确反映真实情况。
- 论文提出了一种因果市场模拟器，通过训练在合成市场先验上，解决了流量与价格关系的混淆问题。
- 实验结果显示，模型在多个市场数据集上表现出高效的因果响应，提升了对元订单影响的估计准确性。
📝 摘要（中文）
当交易者执行大额订单时，关键问题在于“如果我不交易，价格会是多少？”市场影响因此是一个因果量。然而，传统模型（如Almgren-Chriss模型和传播模型）通过流量回归价格，这在信息流影响价格的情况下存在偏差。本文提出了一种概念验证的因果市场模拟器：一种基于连续时间因果先验数据拟合网络（PFN），该网络在一个合成市场先验上进行训练，其中隐藏的信息流通过构造混淆了流量与价格的关系。实验结果表明，PFN能够在高混淆情况下追踪无偏真实参数，而其去干预的模型则回归到窗口前均值。该模型在ABIDES限价单簿中对元订单的响应表现出显著的因果转移证据。
🔬 方法详解
问题定义：本文旨在解决交易者在执行大额订单时，如何准确估计市场影响的问题。现有方法在信息流影响价格的情况下，存在显著的偏差，无法提供准确的市场影响估计。
核心思路：论文提出的因果市场模拟器（PFN）通过在合成市场先验上训练，利用隐藏的信息流来混淆流量与价格的关系，从而实现无偏估计。该设计允许研究者精确控制混淆强度，进而分析市场影响。
技术框架：整体架构包括一个连续时间的因果先验数据拟合网络，模型通过合成市场数据进行训练，包含流量输入和价格输出的关系建模。模型的去干预版本用于验证无偏性。
关键创新：最重要的技术创新在于通过因果建模来处理市场影响，避免了传统方法中的偏差，提供了更为准确的市场影响估计。与现有方法相比，PFN能够在高混淆情况下保持无偏性。
关键设计：模型的关键设计包括对混淆强度的精确控制、损失函数的选择，以及网络结构的优化。通过去干预模型的设计，验证了模型在真实市场数据上的有效性。
🖼️ 关键图片
📊 实验亮点
实验结果显示，因果市场模拟器在ABIDES限价单簿中对元订单的响应达到了真实参数的高效追踪，表现出+0.38的标准差提升。此外，模型在BTC/ETH/SOL永续合约和纳斯达克股票的流量突发检测中，成功复制了因果转移的证据，显著提高了对市场影响的估计准确性。
🎯 应用场景
该研究的潜在应用领域包括金融市场的交易策略优化、风险管理和市场预测等。通过提供更准确的市场影响估计，交易者可以更有效地执行大额订单，降低市场冲击，提高交易效率。未来，该模型有望扩展到其他金融工具和市场环境中。
📄 摘要（原文）
When a trader executes a large order, the relevant question is not \emph{what will the price be?} but \emph{what would the price have been had I not traded?} Market impact is thus a causal quantity, yet classical models (Almgren--Chriss, propagators) regress price on flow, which is biased whenever informed flow moves prices through information as well as trading. We present a proof-of-concept \emph{Causal Market Simulator}: a continuous-time causal prior-data fitted network (PFN) trained on a synthetic market prior in which a hidden informational flow confounds the flow--price relation by construction, with confounding strength an exact per-episode knob. On this prior, the naive estimator's bias follows a closed form we confirm empirically (up to 68\% overestimation), while the PFN tracks the unbiased true-parameter oracle at high confounding. Its \emph{do-ablated twin}, the same model with the intervention inputs removed, collapses onto the pre-window mean. Zero-shot, the model responds at the ground-truth scale to metaorders in an ABIDES limit-order book ($+0.38$ pre-onset standard deviations against seed-replay counterfactuals), and its intervention response replicates on about 2,300 detected flow bursts in BTC/ETH/SOL perpetuals and 3,800 on Nasdaq equities, evidence consistent with, though not proof of, causal transfer. Off-prior the response degrades gracefully, and split-conformal recalibration repairs interval calibration. Transfer hinges on encoding the metaorder as added flow rather than a clamped price level, not on more training. A recurrent state-space variant closes the remaining shape gap: its impact accrues over execution duration where the transformer's stays flat, raising the correlation with measured metaorder impact from $+0.68$ to between $+0.92$ and $+0.96$, at the cost of weaker on-prior deconfounding that traces to the training recipe rather than to recurrence.