信任不可检验的事物：Doubly Robust Models 的验证与诊断

[

Lyft Engineering

Lyft 工程

](https://eng.lyft.com/?source=post_page---publication_nav-25cd379abb8-00853df009df---------------------------------------)

[

Lyft Engineering

](https://eng.lyft.com/?source=post_page---post_publication_sidebar-25cd379abb8-00853df009df---------------------------------------)

Stories from Lyft Engineering.

Lyft Engineering 的故事。

written by Ross Chu and Shima Nassiri

作者 Ross Chu 和 Shima Nassiri

The Causal Frontier: Measurement Beyond Randomization

因果前沿：超越随机化的测量

The gold standard for determining the causal impact of a policy or product change at a company like Lyft is the A/B test (randomized experiment). By randomly assigning users to a treatment or control group, A/B tests inherently eliminate bias, providing clean estimates of the Average Treatment Effect (ATE). However, many critical business questions and large-scale initiatives simply cannot be randomized. This forces scientists to move past traditional experimentation and leverage quasi-experimental methods.

在像 Lyft 这样的公司中，确定政策或产品变更因果影响的金标准是 A/B test（随机化实验）。通过随机地将用户分配到治疗组或对照组，A/B test 天生消除了偏差，提供 Average Treatment Effect (ATE) 的干净估计。然而，许多关键业务问题和大规模举措根本 无法随机化。这迫使科学家们超越传统实验，并利用 准实验 方法。

We rely on non-randomized measurement in several key scenarios across Lyft:

我们在 Lyft 的几个关键场景中依赖非随机化测量：

Partnerships and Policies: Assessing the incremental impact of a partnership (e.g., linking two company accounts) is often a non-randomized assignment. Since these collaborations require coordinated operational work across both companies and are typically announced or promoted broadly, this makes controlled randomization impractical.
Partnerships and Policies: 评估合作伙伴关系（例如，链接两个公司账户）的增量影响通常是非随机分配的。由于这些合作需要两家公司协调的运营工作，并且通常被广泛宣布或推广，这使得控制随机化不切实际。
Long-Term Effect (LTE): Measuring effects that unfold over a lon...