Covariate-adaptive randomization using historical data and bootstrap-based inference

Nodoka Seya; Masataka Taguri; Osamu Togo; Kengo Kamatani

Home/Archive/arXiv

arXiv · 2610.26254

Covariate-adaptive randomization using historical data and bootstrap-based inference

Nodoka Seya·Masataka Taguri·Osamu Togo·Kengo Kamatani

View original source ↗Read source document ↗Export metadata ↓

Abstract

Covariate-adaptive randomization (CAR) can improve the efficiency of average treatment effect (ATE) estimation by reducing chance imbalances in prognostic baseline covariates. A recently proposed CAR method that balances multiple covariates, including continuous ones, can achieve optimal precision when the conditional mean of the outcome can be expressed as a linear combination of the covariate features used for the CAR procedure. However, this condition may be difficult to satisfy when the relationship between covariates and the outcome is nonlinear. Furthermore, directly balancing multiple covariates may reduce estimation precision in finite samples. We propose three CAR methods that incorporate a prognostic score learned from individual-level historical data into their imbalance measures. The prognostic score reduces multiple covariates to one dimension while giving larger weights to covariates that are more strongly associated with the outcome. Another challenge with CAR is difficulty in inference due to dependence between subjects induced by the CAR procedure. Existing bootstrap-based inference methods emulate the CAR design by reassigning treatment within each bootstrap sample, which causes difficulty in interval estimation under CAR with continuous covariates. We propose a bootstrap method with unequal probability resampling corresponding to the CAR design, avoiding reassignment within each bootstrap sample. In simulations, the proposed CAR methods reduced prognostic imbalance, thereby improving estimation precision compared to existing assignment designs, and the proposed bootstrap method accurately assessed estimation uncertainty under our proposed CAR. An application to Clazosentan trials demonstrated that the proposed CAR methods similarly improved prognostic balance, and the proposed bootstrap method closely reproduced the distribution induced by the proposed CAR.

Explore related subjects

stat.ME

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Nodoka Seya, Masataka Taguri, Osamu Togo, Kengo Kamatani. 2026-10-19. Covariate-adaptive randomization using historical data and bootstrap-based inference. https://arxiv.org/abs/2610.26254

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Estimation of Treatment Effects Under Nonstationarity via the Truncated Policy Gradient Estimator

Randomized experiments (or A/B tests) are widely used to evaluate interventions in dynamic systems such as recommendation platforms, marketplaces, and digital health. In these settings, interventions affect both current and future system states, so estimating the global average treatment effect (GATE) requires accounting for temporal dynamics, which is especially challenging in the presence of nonstationarity; existing approaches suffer from high bias, high variance, or both. In this paper, we address this challenge via the novel Truncated Policy Gradient (TPG) estimator, which replaces instantaneous outcomes with short-horizon outcome trajectories. The estimator admits a policy gradient interpretation: it is a truncation of the first-order approximation to the GATE, yielding explicit bias and variance bounds in nonstationary Markovian settings. For a fixed truncation size, we establish a central limit theorem and, under an additional mean-stability condition, the consistency of a variance estimator for nonstationary single-trajectory data. We validate our theory with two real-world case studies. The results show that relative to existing approaches, a well-calibrated TPG estimator can achieve a favorable balance between bias and variance in nonstationary settings, highlighting the value of the policy-gradient perspective for designing effective estimators under complex dynamics.

stat.ME↗

arXiv2026

Weak Factor Analysis via Temporally Weighted PCA

Principal component analysis (PCA) is arguably the most widely used approach for large-dimensional factor analysis. While it is effective when the factors are sufficiently strong, it can be inconsistent when the factors are weak and/or the noise has complex dependence structure. We argue that the inconsistency often stems from bias and introduce a general approach to restore consistency. Specifically, we propose a general weighting scheme for PCA and show that suitable weights can restore consistency for much weaker factors under the dependent-noise structures allowed by our model. Under the stronger signal-to-noise conditions needed for inference, we also establish asymptotic normality of the loading and factor estimators. While an oracle weighting matrix may require knowledge about the factors and covariance of the idiosyncratic noise that are not known a priori, we develop an agnostic approach to adaptively choose from a large class of weighting matrices that can be viewed as PCA for weighted linear combinations of auto-covariances among the observations. Theoretical and numerical results demonstrate the merits of our methodology over the usual PCA and other recently developed techniques for large-dimensional approximate factor models.

stat.ME↗

arXiv2026

RAPSEM: Identifying Latent Mediators Without Sequential Ignorability via a Rank-Preserving Structural Equation Model

Standard structural equation models (SEMs) are often used to identify latent mediators. However, valid inference typically relies on the strong, frequently violated Sequential Ignorability assumption. We introduce the Rank-Preserving Structural Equation Model (RAPSEM), which increases robustness through G-estimation while maintaining the measurement model's integrity through a two-stage method of moments (2SMM) for factor score corrections. RAPSEM replaces the no unmeasured mediator-outcome confounding with the weaker no unobserved effect modification assumption. By leveraging treatment randomization, RAPSEM achieves identification in a manner equivalent to instrumental variable estimation through structurally emerging instruments. Specifically, identification relies on treatment-covariate interactions that influence the mediator but have no direct effect on the outcome, allowing researchers to utilize natural heterogeneity in treatment response as a testable source of identification. We provide a robustness assessment for the core identifying assumption and establish the consistency and asymptotic normality of the resulting estimator. Simulation studies demonstrate that RAPSEM remains unbiased under unobserved confounding, whereas standard SEM yields biased results. RAPSEM achieves reasonable power for sample sizes above 500, depending on the strength of the structural instruments. The method is implemented in the accompanying rapsem R package, and its practical utility is illustrated through an empirical example from educational research. The code is available at https://github.com/PsychometricsMZ/RAPSEM.

stat.ME↗