arXiv subjects

Shuozhi Zuo

Publications and source records attributed to Shuozhi Zuo.

4 recordsLinked to original sources

arXiv2026

Identifiability of the instrumental variable model with the treatment and outcome missing not at random

The instrumental variable model of Imbens and Angrist (1994) and Angrist et al. (1996) identifies the local average treatment effect, also known as the complier average causal effect (CACE). In practice, however, the treatment and outcome are often missing, and when they are missing not at random (MNAR), the CACE is generally not identifiable without further assumptions, because the underlying data distribution itself cannot be recovered. We study when the CACE remains identifiable under MNAR. Through an exhaustive search over missingness mechanisms, we characterize all those that identify the CACE without auxiliary information, in two scenarios: (1) missing data in either the treatment or the outcome alone, and (2) missing data in both the treatment and outcome under prospective data collection. Along the way, we unify existing results and establish many new ones, giving a complete picture of identifiability in each case. Our theory suggests that before any practical data analysis under the instrumental variable model, it is important to check whether the CACE is identifiable under the proposed missingness mechanism; moreover, because the true mechanism is typically unknown and untestable, it is more robust to conduct sensitivity analyses across multiple plausible missingness mechanisms.

stat.ME↗

arXiv2026

Environment-Adaptive Covariate Selection: Learning When to Use Spurious Correlations for Out-of-Distribution Prediction

A common approach to out-of-distribution prediction restricts models to causal or invariant covariates to avoid spurious associations that may change across environments. Despite its theoretical appeal, this strategy can underperform empirical risk minimization when only a subset of the causal parents of the outcome is observed. In such settings, non-causal covariates can serve as proxies for unobserved causal parents and improve prediction when the proxy relationship is stable, but they can hurt when shifts disrupt that relationship. Thus, the optimal covariate set can depend on the specific shift encountered. Because different shifts leave signatures in the unlabeled covariate distribution, we propose an environment-adaptive covariate selection algorithm that maps environment-level summaries to environment-specific covariate sets. These summaries may be hand-crafted or learned from multi-environment data, and prior causal knowledge can be incorporated as constraints. Across simulations and applied datasets, the proposed method improves over static causal, invariant, and other non-adaptive rules under diverse shifts.

stat.ME↗

arXiv2026

Identification and estimation of the conditional average treatment effect with nonignorable missing covariates, treatment, and outcome

Treatment effect heterogeneity is central to policy evaluation, social science, and precision medicine, where interventions can affect individuals differently. In observational studies, covariates, treatment, and outcomes are often only partially observed. When missingness depends on unobserved values (missing not at random; MNAR), standard methods can yield biased estimates of the conditional average treatment effect (CATE). This paper establishes nonparametric identification of the CATE under multivariate MNAR mechanisms that allow covariates, treatment, and outcomes to be MNAR. It also develops nonparametric and parametric estimators and proposes a sensitivity analysis framework for assessing robustness to violations of the missingness assumptions.

stat.ME↗

arXiv2023

Mediation analysis with the mediator and outcome missing not at random

Mediation analysis is widely used for investigating direct and indirect causal pathways through which an effect arises. However, many mediation analysis studies are challenged by missingness in the mediator and outcome. In general, when the mediator and outcome are missing not at random, the direct and indirect effects are not identifiable without further assumptions. In this work, we study the identifiability of the direct and indirect effects under some interpretable mechanisms that allow for missing not at random in the mediator and outcome. We evaluate the performance of statistical inference under those mechanisms through simulation studies and illustrate the proposed methods via the National Job Corps Study.

stat.ME↗