Forecasting Associations, Not Interventions: A Preregistered Causal Audit of Covariate-Aware Time Series Foundation Models
作者: Zong-Han Bai
分类: cs.LG
发布日期: 2026-10-22
备注: 21 pages, 11 figures, 12 tables
💡 一句话要点
提出因变量预测模型以解决干预效应识别问题
🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 时间序列预测 因果推断 干预效应 模型训练 观察性数据
📋 核心要点
- 现有的时间序列预测模型在处理干预性查询时，往往依赖观察性数据，导致无法准确识别干预效应与关联效应的区别。
- 本文提出了一种新的方法，通过对模型进行干预后训练，显著提高了模型在面对反事实查询时的预测准确性。
- 实验结果显示，Chronos-2和TiRex-2等模型在经过干预后训练后，预测误差减少了83%，并在多个基准测试中表现优异。
📝 摘要（中文）
本文探讨了协变量感知时间序列基础模型（TSFMs）在处理干预性查询时的局限性，重点分析了模型在观察性数据下的表现。研究表明，Bayes-optimal观察预测器在没有调整集的情况下，无法准确回答干预查询。通过对Chronos-2、TiRex-2等模型的实证分析，发现这些模型在面对真实的反事实基础真相时，能够有效追踪模型效应的真实差异。进一步的实验表明，经过短时间的干预后训练，模型的预测能力显著提升，减少了干预效应的估计误差。
🔬 方法详解
问题定义：本文旨在解决协变量感知时间序列基础模型在干预查询中的局限性，现有方法在没有调整集的情况下，无法准确识别干预效应与关联效应的差异。
核心思路：论文提出通过对模型进行短时间的干预后训练，来提升模型在反事实查询中的预测能力，从而缩小模型的估计误差。
技术框架：研究采用了Bayes-optimal观察预测器作为理论基础，结合线性最小二乘法和模型的嵌入表示，构建了一个多层次的实验框架，评估模型在不同条件下的表现。
关键创新：最重要的创新在于提出了在观察性数据下，如何通过干预后训练来减少模型的混淆效应，并且在多个模型中验证了这一方法的有效性。
关键设计：在实验中，采用了精确的反事实基础真相作为对照，设置了多个参数以评估模型的表现，并通过低秩适应技术优化了模型的训练过程。
📊 实验亮点
实验结果显示，Chronos-2、TiRex-2等模型在面对真实反事实基础真相时，能够有效追踪模型效应的真实差异，R²值达到0.90-1.00。经过干预后训练，模型的预测误差减少了83%，并在DoTime基准测试中表现出色，提升幅度显著。
🎯 应用场景
该研究的潜在应用领域包括经济学、公共政策、医疗健康等领域，尤其是在需要进行因果推断和干预效果评估的场景中。通过提高模型的预测准确性，能够为决策者提供更可靠的依据，从而优化资源配置和政策制定。
📄 摘要（原文）
Covariate-aware time series foundation models (TSFMs) accept a future path for a control variable and return a forecast, inviting what-if queries. The query is interventional; the models are trained on observational data. We determine whether their answers are the conditional slope $\partial E[Y|X]$ or the interventional effect $\partial E[Y|do(X)]$, at three levels. Theory: a Bayes-optimal observational predictor answers interventional queries associationally unless the adjustment set is available and used; for in-context least squares we derive the resulting confounding gap -- a model's effect estimate under an observational minus under a randomized context -- in closed form, a parameter-free reference line. Behaviour: on a preregistered grid with exact counterfactual ground truth, Chronos-2, TiRex-2, TabPFN-TS and TimesFM-2.5 all track this line ($R^2$ 0.90-1.00) and respond more confounded than regression itself (slopes 1.3-1.8), on linear and nonlinear structural models. Pairs of models with identical observational laws prove this is not an identifiability limit: when the effect is identifiable, least squares recovers 100% of the true effect difference, Chronos-2 75%, TiRex-2 34%. Representation: linear probes decode the data-generating regime (94-96%) and the true effect ($R^2$ 0.7-0.9) from embeddings that the forecasts ignore. On two physical systems with randomized actuators, pretrained models under-respond by 2-5x and ignore in-window confounding. Minutes of interventional post-training on 3,000 synthetic series cut the gap by up to 83% and transfer outside the training prior. Its forecasting cost is a replay-distribution effect: replaying the model's own pretraining prior cuts it from +47% to +10% MASE, and with low-rank adaptation yields a recipe meeting the preregistered criterion over three seeds (52-61% gap reduction at +1-3% cost) while improving the DoTime benchmark.