OCIS · Fall 2021¶
- 共 13 场 · 8 篇精读
本季导览¶
自动生成：归纳本季主线与值得先看的几场，不打分、不排名。
OCIS Fall 2021 的 13 场报告可归纳为四条主线：图模型与反事实框架的统一（Richardson）、非标准数据与因果推断（Wang, Westling）、高效率估计与双机器学习（Tian, Athey & Wager）、敏感性分析与稳健推断（Cinelli, Fogarty, Lee, Li, Antonelli）。此外，工具变量与孟德尔随机化（Lee, Sanderson）、因果发现与可解释性（Malinsky）、软干预泛化（Correa & Gnecco）构成独立支线。
最突出的主线是图模型与反事实框架的统一。Richardson 的 SWIG 为 DAG 与潜在结果提供了简洁的图形翻译，直接回应了 NPSEM 和孪生网络的过度假设问题。这条线在非标准数据方向得到延伸：Wang 将潜在结果框架推广到分布值结果，利用 Wasserstein 几何定义因果效应；Westling 则针对连续暴露构造非参数检验，其 EIF 构造与 Richardson 的图形识别形成互补。
高效率估计与双机器学习是另一条密集主线。Tian 系统性地为任意可识别因果效应（由 ID 算法输出）构造 DML 估计器，解决了后门/前门之外复杂表达式的双重稳健估计问题。Athey & Wager 的教程则聚焦异质性处理效应（CATE）的 ML 估计与验证，与 Tian 的通用框架形成从“识别-估计”到“异质性推断”的完整链条。敏感性分析方面，Cinelli 提供了基于偏 R² 的透明框架，与 Lee 的证据因子分析（利用多个可能无效的 IV 提取正交证据）共享“量化假设违背影响”的核心追问。
对于快速入门，建议按以下路径：基础：Richardson（SWIG 统一框架）打底，Cinelli（敏感性分析）建立稳健性思维；进阶：Tian（DML 通用估计）与 Athey & Wager（CATE 估计）构成估计主线；前沿：Wang（分布值结果）与 Westling（连续暴露检验）展示非标准数据扩展；工具变量：Lee（证据因子）与 Sanderson（多时间点 MR）提供 IV 方向的两种新视角。
报告列表¶
- Single World Intervention Graphs: A simple framework for unifying graphs and potential outcomes with applications to mediation analysis¶
讲者: Thomas Richardson · 讨论人: Mats Stensrud · 2021-11-30
链接：视频 · 幻灯片
摘要
Causal models based on potential outcomes, also known as counterfactuals, were introduced by Neyman (1923) and later popularized by Rubin. Causal Directed Acyclic Graphs (DAGs) are another approach, originally introduced by Wright (1921), but subsequently significantly generalized and extended by Spirtes and Pearl among others. In this talk I will first present a simple approach to unifying these two approaches via Single-World Intervention Graphs (SWIGs). The SWIG encodes the counterfactual ind…- Causal inference on distribution functions¶
讲者: Linbo Wang · 讨论人: Hongtu Zhu · 2021-11-16
链接：视频 · 幻灯片 · arXiv
摘要
Understanding causal relationships is one of the most important goals of modern science. So far, the causal inference literature has focused almost exclusively on outcomes coming from the Euclidean space Rp. However, it is increasingly common that complex datasets collected through electronic sources, such as wearable devices, cannot be represented as data points from Rp. In this paper, we present a novel framework of causal effects for outcomes from the Wasserstein space of cumulative distribut…- Estimating Identifiable Causal Effects through Double Machine Learning - Graph-based & Data-driven Approaches¶
讲者: Jin Tian · 讨论人: Ilya Shpitser · 2021-11-09
链接：视频 · 幻灯片
摘要
Inferring causal effects from observational data is a fundamental task throughout the empirical sciences. General methods have been developed to decide the identifiability of a target effect from a combination of observational data and the causal graph underlying the system. In practice, however, there are still challenges to estimating identifiable causal functionals from finite samples. We aim to fill this gap between causal identification and causal estimation. In this talk, I will discuss tw…- Randomization Inference beyond the Sharp Null: Bounded Null Hypotheses and Quantiles of Individual Treatment Effects （暂无精读）¶
讲者: Xinran Li · 讨论人: Panos Toulis · 2021-11-02
链接：视频 · 幻灯片
摘要
Randomization (a.k.a. permutation) inference is typically interpreted as testing Fisher's "sharp" null hypothesis that all effects are exactly zero. This hypothesis is often criticized as uninteresting and implausible. We show, however, that many randomization tests are also valid for a "bounded" null hypothesis under which effects are all negative (or positive) for all units but otherwise heterogeneous. The bounded null is closely related to important concepts such as monotonicity and Pareto ef…- Transparent and Robust Causal Inference in the Social and Health Sciences¶
讲者: Carlos Cinelli · 讨论人: Guido Imbens · 2021-10-26
链接：视频 · 幻灯片
摘要
The past few decades have witnessed rapid and unprecedented theoretical progress on the science of causal inference, ranging from the "credibility revolution” with the popularization of quasi-experimental designs, to the development of a complete solution to non-parametric identification with causal graphical models. Most of this theoretical progress, however, relies on strong, exact assumptions, such as the absence of unobserved common causes (ignorability assumptions), or the absence of certai…Columbia University & Universidad Autónoma de Manizales ) & Nicola Gnecco (University of Geneva) Talk1: Generalizing the Effect of Soft Interventions [ Video （暂无精读）¶
讲者: Juan Correa ( · 2021-10-19
链接：视频 · 幻灯片
- Prepivoting in Finite Population Causal Inference¶
讲者: Colin Fogarty · 讨论人: Tirthanker Dasgupta - · 2021-10-12
链接：视频 · 幻灯片
摘要
In finite population causal inference exact randomization tests can be constructed for sharp null hypotheses, hypotheses which fully impute the missing potential outcomes. Oftentimes inference is instead desired for the weak null that the sample average of the treatment effects takes on a particular value while leaving the subject-specific treatment effects unspecified. Without proper care, tests valid for sharp null hypotheses may be anti-conservative even asymptotically should only the weak nu…- Estimation of causal effects of an exposure at multiple time points through Multivariable Mendelian randomization （暂无精读）¶
讲者: Eleanor Sanderson · 讨论人: Stephen Burgess · 2021-10-05
链接：视频 · 幻灯片
摘要
Mendelian Randomisation (MR) is a powerful tool in epidemiology which can be used to estimate the causal effect of an exposure on an outcome in the presence of unobserved confounding, by utilising genetic variants as instrumental variables (IVs) for the exposure. The effects obtained from MR studies are often interpreted as the lifetime effect of the exposure in question. However, the causal effects of many exposures are thought to vary throughout an individual’s lifetime and there may be period…- Evidence factors from multiple, possibly invalid, instrumental variables¶
讲者: Youjin Lee · 讨论人: Jose Zubizarreta · 2021-09-28
链接：视频 · 幻灯片
摘要
Instrumental variables have been widely used to estimate the causal effect of a treatment on an outcome in the presence of unmeasured confounders. When several instrumental variables are available and the instruments are subject to possible biases that do not completely overlap, a careful analysis based on these several instruments can produce orthogonal pieces of evidence (i.e., evidence factors) that would strengthen causal conclusions when combined. We develop several strategies, including st…Nonparametric tests of the causal null with non-discrete exposures¶
讲者: Ted Westling · 讨论人: Oliver Dukes · 2021-09-21
链接：视频 · 幻灯片 · arXiv
摘要
Many methods have been developed to test for the presence of a causal effect of a discrete exposure on an outcome when there are no unobserved confounders. In this talk, we introduce a class of nonparametric tests of the null hypothesis that there is no average causal effect of an arbitrary univariate exposure on an outcome when there are no unobserved confounders. Our tests apply to discrete, continuous, and mixed discrete-continuous exposures. We demonstrate that our proposed tests are doubly-…Explaining the Behavior of Black-Box Prediction Algorithms with Causal Learning¶
讲者: Daniel Malinsky · 讨论人: Joshua Loftus · 2021-09-14
链接：视频 · 幻灯片 · arXiv
摘要
We propose to explain the behavior of black-box prediction methods (e.g., deep neural networks trained on image pixel data) using causal graphical models. Specifically, we explore learning the structure of a causal graph where the nodes represent prediction outcomes along with a set of macro-level “interpretable” features, while allowing for arbitrary unmeasured confounding among these variables. The resulting graph may indicate which of the interpretable features, if any, are possible causes of…Heterogeneous causal effects of neighborhood policing in New York City with staggered adoption of the policy （暂无精读）¶
讲者: Joseph Antonelli · 讨论人: Matthew Cefalu · 2021-09-07
链接：视频 · 幻灯片 · arXiv
摘要
Communities often self select into implementing a regulatory policy, and adopt the policy at different time points. In New York City, neighborhood policing was adopted at the police precinct level over the years 2015-2018, and it is of interest to both (1) evaluate the impact of the policy, and (2) understand what types of communities are most impacted by the policy, raising questions of heterogeneous treatment effects. We develop novel statistical approaches that are robust to unmeasured confou…Estimating heterogeneous treatment effects in R （暂无精读）¶
讲者: Susan Athey and Stefan Wager · 2021-08-31
链接：视频 · 幻灯片
摘要
This tutorial will survey recent advances in machine learning based estimation of conditional average treatment effects under unconfoundedness. We will also discuss methods for validating and interpreting estimates of treatment heterogeneity. Methods will be illustrated using numerical examples in R.Maintained by 陈星宇 · Homepage · Source