spatial difference-in-differences

This study addresses estimation bias in design-based inference under spatial interference, arising from unobserved outcomes and model misspecification. By integrating a design-based inference framework with spatial outcome modeling, we elucidate the intrinsic mechanisms through which unbiased estimators become biased. Numerical simulations and a reanalysis of hot-spot policing experiments demonstrate that this bias diminishes solely as observation density increases, while model-based standard errors converge to their theoretical optimum. These findings clarify the impact of modeling error on spatial causal inference, providing essential theoretical foundations and practical guidance for optimizing the evaluation of spatially correlated experiments.

Geospatial impact evaluations often grapple with ambiguity in defining the exposure units, timing, and intensity of interventions, particularly when multiple plausible exposure definitions exist. This study introduces the concept of “treatment geometry” as a foundational framework to systematically characterize the spatiotemporal footprint of interventions derived from Earth observation data. Centered on key trade-offs—including spatial resolution, temporal alignment, spillover effects, and boundary uncertainty—the framework provides diagnostic tools that enable researchers to identify which geometric definition choices are most critical for causal identification, rather than defaulting to a single methodological approach. Empirical applications to air pollution, wildfires, and forest policy demonstrate that this approach substantially enhances the credibility of causal inference and the rigor of empirical design.

This paper addresses the challenge of identifying structural break points—termed “treatment effect boundaries”—where causal effects abruptly vanish across spatiotemporal dimensions. Methodologically, it introduces the first unified theoretical framework that jointly models spatial and temporal boundaries, defines structural parameters governed by shared dynamical systems, establishes rigorous identifiability conditions, and develops consistent, asymptotically normal estimators. Integrating information diffusion modeling with causal inference theory, the approach is validated via Monte Carlo simulations under heterogeneous spatiotemporal data, demonstrating robustness and statistical efficiency. Key contributions are: (1) a formal characterization of the critical threshold at which policy interventions transition from locally effective to systemically ineffective; and (2) practical, implementable tools for boundary detection and estimation, thereby providing both theoretical foundations and empirical support for institutional transition analysis and precision policy design.

This paper addresses the failure of causal identification in longitudinal panel data due to spatiotemporal interference—where an individual’s outcome is affected by others’ past treatment assignments. We propose a design-based causal inference framework that, under minimal assumptions (unknown interference structure and sequential ignorability), formally defines and identifies separable direct effects and spatiotemporal spillover effects for the first time. We demonstrate that conventional fixed-effects and difference-in-differences (DID) estimators suffer from systematic bias under interference. To overcome this, we construct a new estimator with consistency and asymptotic normality. Theoretical analysis, Monte Carlo simulations, and replications of two canonical empirical studies validate our approach: it substantially reduces estimation bias in spillover effects and effectively corrects the failure of standard panel methods under complex interference patterns.

研究设计了基于Matérn点过程的空间干预方案，通过调整干预点间的最小距离r来平衡偏差与方差，以估计无污染的平均干预效应。

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.

This study addresses causal inference for continuous spatiotemporal point processes subject to spillover and lingering effects under unit-level interventions. It proposes the first causal inference framework grounded in the potential outcomes paradigm, modeling the observed post-intervention process as an unlabeled superposition of control and treated components. Identification is achieved by separately analyzing regions within and outside the support of the intervention, leveraging a structured point process model to recover causal contrasts in non-support areas. Estimation employs a likelihood-based stochastic EM algorithm, augmented with a predictable block-wise hard EM surrogate, making it applicable to history-dependent processes such as Poisson and Hawkes processes. The method provides non-asymptotic error bounds and plug-in inference guarantees. Empirical validation on wastewater injection and seismicity data from Oklahoma demonstrates its practical efficacy.

This study addresses the inferential bias in the average treatment effect on the treated (ATT) arising from data-driven control group selection and covariance matrix estimation within difference-in-differences frameworks. To mitigate this, it proposes a selective inference framework that extends exact Gaussian procedures to settings with unknown covariance by employing plug-in estimators to correct for selection bias, applicable to both individual panel and repeated cross-sectional data. This work unifies inference under complex scenarios involving staggered adoption, treatment effect heterogeneity, and sample reuse. Furthermore, it establishes uniform conditional and marginal coverage guarantees, proves the asymptotic equivalence between plug-in estimators and known-covariance confidence intervals, and derives convergence rates for interval lengths, thereby ensuring the statistical validity of ATT confidence intervals.

This study addresses the lack of systematic criteria for covariate selection in difference-in-differences (DID) designs, which often leads to arbitrary specifications of the conditional parallel trends assumption. It introduces causal graphical models to formally identify the set of covariates required to satisfy this assumption, highlighting the critical role of time-invariant covariates and distinguishing between treatment types and implementation strategies to clarify treatment–confounding feedback mechanisms. By integrating a multi-period DID framework with conditional independence tests and consistency analysis of adjustment sets, the paper proposes principled criteria for covariate inclusion, resolving the prevalent mismatch between adjustment sets and estimation methods. This approach substantially enhances the identification validity and robustness of causal effect estimates in DID analyses.

Existing causal models struggle to distinguish between the immediate and persistent effects of interventions in time-dynamic systems, particularly when such interventions alter the system’s equilibrium behavior. This work proposes a novel paradigm grounded in system and state representations, integrating causal directed acyclic graphs, the potential outcomes framework, and dynamic systems theory. By introducing an equilibrium-state assumption and employing state-space modeling, the study reformulates the causal inference framework to better capture temporal dynamics. It innovatively defines an equilibrium-oriented “zero effect” concept and combines it with a strategic selection of time points to enable valid identification of time-varying causal parameters. The approach establishes clear criteria for categorizing causal effects under dynamic interventions, substantially enhancing the interpretability and practical utility of causal inference in equilibrium analysis.