Market Impact Estimation via Causal Foundation Models

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.

当交易者执行大额订单时，相关的问题不是“价格会是多少？”，而是“如果我没有进行交易，价格会是多少？”因此，市场影响是一个因果量，然而经典模型（如 Almgren—Chriss 模型、传播算子）通过流量对价格进行回归，只要知情流量既通过信息又通过交易来推动价格，这种回归就会产生偏差。

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.

我们提出了一个概念验证型的“因果市场模拟器”：这是一个连续时间的因果先验数据拟合网络（PFN），它在合成市场先验上进行训练，其中隐藏的信息流通过构建方式混淆了流量与价格之间的关系，且混淆强度在每个片段中都是一个精确可调的旋钮。

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.

在此先验条件下，朴素估计量的偏差遵循我们通过实证确认的闭式解（高达 68% 的高估），而 PFN 在高混淆情况下仍能追踪无偏的真实参数预言机。其“do-消融孪生模型”（即移除了干预输入的相同模型）则会坍缩至窗口前均值。

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.

在零样本情况下，该模型对 ABIDES 限价订单簿中的元订单（metaorders）做出了符合真实规模的响应（相对于种子重放反事实，偏离了 +0.38 个前置标准差），其干预响应在 BTC/ETH/SOL 永续合约中检测到的约 2,300 次流量爆发以及纳斯达克股票中的 3,800 次流量爆发中得到了复现，这些证据与因果迁移一致，尽管尚不能作为其证明。