counterfactual analysis

Counterfactual prediction under evolving intervention policies or hypothetical decision scenarios remains challenging due to unobservable potential outcomes, hindering model identifiability, evaluation, and generalization. Method: We propose the first systematic theoretical framework addressing this challenge—comprising (i) identifiability conditions for counterfactual prediction models, (ii) a performance evaluation system targeting loss, AUC, and calibration, and (iii) robust hyperparameter selection under model misspecification. Our approach integrates causal inference principles, doubly robust estimation, and loss-driven evaluation metric design. Contribution/Results: Validated via simulation studies and a real-world clinical application—cardiovascular risk prediction in statin-naïve populations—the framework significantly improves out-of-distribution generalization and clinical decision reliability in counterfactual settings.

This study addresses the limitations of traditional control-based causal inference methods—such as matching and difference-in-differences—in settings characterized by pervasive or structurally ambiguous spillover effects, where reliance on uncontaminated control units impedes accurate identification of both average direct and spillover effects. Within the potential outcomes framework, this work provides the first systematic comparison between control-based and prediction-based counterfactual approaches—including interrupted time series and machine learning control—in terms of their identification capabilities. Through simulation and empirical analyses, the authors demonstrate that in environments with widespread interference, prediction-based methods can more reliably estimate certain causal parameters over short horizons, circumventing the stringent assumption of unperturbed units and thereby offering a promising alternative for causal inference under complex interference.

Traditional counterfactual reasoning relies on acyclic structural causal models (SCMs), limiting its applicability to real-world systems with feedback loops—e.g., biological regulatory networks. This work extends counterfactual inference to **general cyclic SCMs**, focusing on **shift-scale soft interventions**, i.e., differentiable translations and scalings of mechanism functions. We propose a computational framework based on implicit function differentiation and differentiable optimization, enabling stable solution of nonlinear equation systems induced by mechanism transformations. Our method yields differentiable and consistent estimation of counterfactual distributions in cyclic systems. Experiments demonstrate its effectiveness and robustness on both synthetic cyclic SCMs and real biological pathways. To our knowledge, this is the first theoretically sound and computationally tractable counterfactual analysis tool for complex systems exhibiting strong feedback dynamics.

Traditional causal inference is constrained by a binary paradigm—“observation vs. intervention”—which fails to model intermediate intervention intensities and lacks a continuous quantification framework for causal effects. Method: We propose a continuous causal modeling paradigm, introducing the concept of “causal geodesics”: shortest smooth paths from observational to interventional distributions in a probability distribution metric space. Counterfactual effects are defined via pathwise differentiation along these geodesics, and path modeling and effect estimation are realized through distributional interpolation and tools from differential geometry. Contribution: This work establishes, for the first time, a continuous causal effect quantification framework bridging observability and intervenability. It enables interpretable modeling and estimation of causal effects under arbitrary (including partial or graded) interventions, thereby relaxing the standard discrete-intervention assumption. The framework provides a novel theoretical foundation and computational pathway for causal discovery, incremental policy evaluation, and robust causal inference.

This paper addresses individual-time-point-level counterfactual inference under adaptive treatment strategies in multi-unit, multi-period sequential experiments, aiming to relax strong prior assumptions on intervention policy structure. We propose a nonparametric latent factor model that unifies nonlinear mixed-effects and bilinear factor models. Integrating nonparametric nearest-neighbor estimation with sequential experimental design, we derive the first non-asymptotic, high-probability error bound for individual-time-point-level counterfactual means. We establish theoretical consistency of the estimator and asymptotic validity of associated confidence intervals. The method is validated via simulations and the HeartSteps mobile health clinical trial, demonstrating both statistical accuracy and practical utility. Our core contribution lies in breaking the traditional reliance on restrictive parametric or structural assumptions about treatment policies—enabling high-precision, assumption-light, fine-grained causal inference at the individual-time-point level.

This study addresses a critical limitation in existing design-based simulations used to evaluate inference methods, which often overstate bias induced by spatial correlation due to unrealistic data-generating mechanisms. In particular, share-shift designs that fix outcomes and resample shocks conflate true treatment effects with error dependence structures, leading to misleading assessments. To remedy this, the paper proposes an improved simulation framework that more accurately models error dependence and avoids spurious entanglement between treatment effects and error terms, thereby better approximating real-world data-generating processes. Integrating resampling techniques with share-shift analysis, the proposed approach substantially enhances the reliability of inference evaluation across multiple empirical applications, underscoring the essential role of aligning simulation designs with genuine underlying mechanisms for valid inference assessment.

This study addresses widespread misconceptions in the practical application of the synthetic control method, particularly concerning its reliance on covariates, claims of robustness, and prevailing model selection criteria—assertions often lacking empirical validation and potentially undermining causal inference reliability. Through rigorous theoretical analysis and extensive simulation experiments, the paper systematically evaluates these common misunderstandings and compares the performance of standard implementations against alternative approaches. The findings uncover critical pitfalls in current practices and, grounded in empirical evidence, offer concrete recommendations for more principled implementation and interpretation. By doing so, the work provides researchers with a practical guide to significantly enhance the quality and credibility of causal inferences derived from synthetic control methods.

In high-stakes domains such as healthcare and criminal justice, it is often infeasible to re-conduct randomized controlled trials (RCTs) after updating machine learning models, rendering the causal effects of these updates on downstream outcomes—such as patient survival or recidivism rates—difficult to assess. This work proposes a novel partial identification approach that leverages historical RCT data and fine-grained relationships between prediction accuracy and downstream outcomes. Under two monotonicity assumptions—individual-level “counterfactual correctness” (i.e., correct predictions never lead to worse outcomes) and a trust relationship between subgroup predictive performance and outcomes—the method constructs tight bounds on the causal effect of the updated model. Simulations demonstrate that this approach yields more informative causal effect estimates compared to existing techniques.

This study addresses the challenge of causal inference in N-of-1 behavioral health case studies, where unobserved confounding impedes valid estimation. The authors propose the Ω causal estimator, which achieves identification without measuring or adjusting for confounders by leveraging functional contrasts over the support set of the outcome variable, requiring only the positivity assumption. This approach pioneers a support-based—rather than distribution-based—framework for causal inference, integrating de Finetti’s subjective probability interpretation with a theory of intervention–observation support consistency. A recall-baseline substitution mechanism bridges support-level contrasts to mean-level causal effects. The method’s feasibility is demonstrated in a case study on cognitive behavioral therapy for anxiety, offering clinicians a practical and robust tool for individualized causal inference.