arXiv subjects

Zhenke Wu

Publications and source records attributed to Zhenke Wu.

At least 19 recordsLinked to original sources

arXiv2026

A Distribution Mapping Approach to Counterfactually Fair Reinforcement Learning

Reinforcement learning (RL) seeks to optimize sequential decisions to maximize population-level benefits over time. However, when deployed in high-stakes settings such as healthcare, RL decisions might systematically restrict some subpopulation's access to valuable services in a manner contrary to the values and goals of stakeholders. Counterfactual fairness (CF) offers a promising framework to address this problem based on causal reasoning. This paper develops a data preprocessing algorithm that, when used in tandem with policy learning, enables CF in RL. Our algorithm relies on a novel quantile distribution mapping method for sequentially estimating the counterfactual states and rewards in the data preprocessing step, subsuming common additivity assumptions used for counterfactual prediction as a special case. We theoretically prove that the per-step level of counterfactual unfairness and infinite-horizon suboptimality gap can be bounded under mild regularity conditions. We also empirically test our algorithm in numerical experiments as well as in application to a real-world interventional digital health dataset.

stat.ML↗

arXiv2026

Bayesian Joint Modeling of Longitudinal Symptomatology Scale Responses and Fall Outcomes via Heterogeneous Latent Transition Analysis

The Study of Women's Health Across the Nation (SWAN) has followed women for over 30 years, from midlife premenopause until later life. The study has 16 surveys at approximately 2 years intervals that cover a wide range of physical and psychological symptoms. These multivariate categorical survey responses potentially contain rich health-related information. Temporal trajectories of the survey responses can be characterized by both the responses profiles and the evolving dynamics of the responses over time. To capture those two features and investigate how they inform subsequent health outcomes, we propose a joint multi-layer latent transition model. We combine a latent transition model that classifies individuals based on their response profiles over time with an additional layer of clustering of these latent class transition sequences, with the goal of connecting these cluster profiles with health outcomes: in this application, self-reported falls. In addition, we evaluate the operating characteristics of the method through simulation studies.

stat.AP↗

arXiv2026

Dynamic Prediction of Alternating Recurrent Events via Neural Network

Alternating recurrent events -- event-times of a specific nature that trigger a secondary refractory period -- occur in a wide-range of fields, including behavioral science, criminal justice, and biostatistics. Analysis of these events requires careful attention to the statistical nuance, including correlated observations and repeated outcomes subject to potential censoring. We develop an online dynamic prediction framework appropriate for predicting subsequent alternating recurrent events, by developing neural network theory for a statistical audiences and applying inverse probability weighted pseudo-observations. The proposed model is applied to dynamically predict alternating recurrent event-free time, showing good performance in simulation, and outstanding capability in application to predicting periods of low mood for first-year medical residents. We close with a discussion.

stat.ML↗

arXiv2026

Neural Networks as Linear Regression: An Introduction for Statisticians

Neural networks are a commonly used prediction tool in computer science and statistics. However, the barrier to entry of this interesting field remains high, particularly for classical statisticians trained in a frequentist perspective. In this letter, we demystify neural networks by describing networks that approximate a linear regression and describe common customizations that provide a foundation for further study.

stat.ML↗

arXiv2026

Sparse Longitudinal Functional Principal Component Analysis for Episodic Ambulatory Behavioral Assessments

Accurately monitoring mental fatigue is critical for improving workplace safety and productivity. A recent study examined unobtrusively collected smartphone typing speed as a potential ambulatory proxy assessment of mental fatigue using data from the Intern Health Study (IHS). While population-level average typing speed patterns were found to be consistent with validated measures of mental fatigue, how these trajectories vary across participants and days may inform opportune moments for just-in-time interventions and remains an open question. Treating typing speed trajectories as sparsely observed functional data, we propose a novel sparse longitudinal functional principal component analysis (sparse LFPCA) method for decomposing variability and predicting individual curves. Specifically, sparse data are accommodated by casting covariance estimation as a structured penalized spline regression problem, enabling simultaneous estimation and smoothing of multiple covariance components while borrowing information across locations in the functional domain. Simulations show that sparse LFPCA (1) accurately estimates eigenfunctions and generates reasonable predictions for underlying curves, and (2) achieves similar or superior performance compared to existing alternatives. Our analysis of typing speed data collected from IHS reveals new and interpretable participant- and day-level patterns not captured by previous analyses and can be used to tailor behavioral interventions.

stat.ME↗

arXiv2026

Prediction-based Inference in Electronic Health Record (EHR)-linked Biobanks with Clinically Informative Outcomes

Electronic health record (EHR)-linked biobank data hold tremendous promise for large-scale discoveries via genome-wide association study (GWAS) on diverse phenotypic traits and biomarkers routinely captured in the EHR. However, heterogeneous missingness in biomarkers compromises the validity and efficiency of statistical analyses. Prediction-based (PB) inference methods meet this challenge by using external machine learning (ML) predictions to impute missing biomarker outcomes, thereby improving statistical power and estimation accuracy in association analyses. Yet, their suitability remains unclear when outcomes are subject to clinically informative observation processes, that is, when laboratory tests are ordered based on both measured and unmeasured patient- and health system-level characteristics. In this paper, we review the statistical underpinnings of popular PB methods and then evaluate nine methods, including four PB methods and five traditional missing-data approaches, under an encompassing set of outcome observation processes for continuous and binary outcomes. PB methods can substantially improve statistical power and estimation efficiency when the missing-data mechanism is correctly specified. Under misspecification, however, these gains require both conditional independence between the covariates of interest and the missingness mechanism and independence between imputation error and the missingness mechanism. Using All of Us (AoU) data, we perform GWAS of six laboratory biomarkers and demonstrate that PB methods can replicate known genetic associations while improving efficiency relative to (weighted) complete-case analysis (CCA). Their performance in replicating existing GWAS results in AoU also depends on imputation quality and the underlying missingness mechanism.

stat.AP↗

arXiv2026

Spatially Robust Inference with Predicted and Missing at Random Labels

When outcome data are expensive or onerous to collect, scientists increasingly substitute predictions from machine learning and AI models for unlabeled cases, a process which has consequences for downstream statistical inference. While recent methods provide valid uncertainty quantification under independent sampling, real-world applications involve missing at random (MAR) labeling and spatial dependence. For inference in this setting, we propose a doubly robust estimator with cross-fit nuisances. We show that cross-fitting induces fold-level correlation that distorts spatial variance estimators, producing unstable or overly conservative confidence intervals. To address this, we propose a jackknife spatial heteroscedasticity and autocorrelation consistent (HAC) variance correction that separates spatial dependence from fold-induced noise. Under standard identification and dependence conditions, the resulting intervals are asymptotically valid. Simulations and benchmark datasets show substantial improvement in finite-sample calibration, particularly under MAR labeling and clustered sampling.

stat.ML↗

arXiv2025

Using Subject-Level Variability to Predict Time-Varying Outcomes: Investigating the Association between Hormone Variability and BMD Trajectories over the Menopausal Transition

Women are at increased risk of bone loss during the menopausal transition; in fact, nearly 50\% of women's lifetime bone loss occurs during this time. The longitudinal relationships between estradiol (E2) and follicle-stimulating hormone (FSH), two hormones that change have characteristic changes during the menopausal transition, and bone health outcomes are complex. However, in addition to level and rate of change in E2 and FSH, variability in these hormones across the menopausal transition may be an important predictor of bone health, but this question has yet to be well explored. We introduce a joint model that characterizes individual mean estradiol (E2) trajectories and the individual residual variances and links these variances to bone health trajectories. In our application, we found that higher FSH variability was associated with declines in bone mineral density (BMD) before menopause, but this association was moderated over time after the menopausal transition. Additionally, higher mean E2, but not E2 variability, was associated with slower decreases in during the menopausal transition. We also include a simulation study that shows that naive two-stage methods often fail to propagate uncertainty in the individual-level variance estimates, resulting in estimation bias and invalid interval coverage.

stat.ME↗

arXiv2025

A Unified Framework for Inference with General Missingness Patterns and Machine Learning Imputation

Pre-trained machine learning (ML) predictions have been increasingly used to complement incomplete data to enable downstream scientific inquiries, but their naive integration risks biased inferences. Recently, multiple methods have been developed to provide valid inference with ML imputations regardless of prediction quality and to enhance efficiency relative to complete-case analyses. However, existing approaches are often limited to missing outcomes under a missing-completely-at-random (MCAR) assumption, failing to handle general missingness patterns (missing in both the outcome and exposures) under the more realistic missing-at-random (MAR) assumption. This paper develops a novel method that delivers a valid statistical inference framework for general Z-estimation problems using ML imputations under the MAR assumption and for general missingness patterns. The core technical idea is to stratify observations by distinct missingness patterns and construct an estimator by appropriately weighting and aggregating pattern-specific information through a masking-and-imputation procedure on the complete cases. We provide theoretical guarantees of asymptotic normality of the proposed estimator and efficiency dominance over weighted complete-case analyses. Practically, the method affords simple implementations by leveraging existing weighted complete-case analysis software. Extensive simulations are carried out to validate theoretical results. A real data example is provided to further illustrate the practical utility of the proposed method. The paper concludes with a brief discussion on practical implications, limitations, and potential future directions.

stat.ME↗

arXiv2025

A Random Forest Inverse Probability Weighted Pseudo-Observation Framework for Alternating Recurrent Events

Alternating recurrent events, where subjects experience two potentially correlated event types over time, are common in healthcare, social, and behavioral studies. Often there is a primary event of interest that, when triggered, initiates a period of treatment and recovery measured via a secondary time-to-event. For example, cancer patients can experience repeated blood clotting emergencies that require hospitalization followed by discharge, people with alcohol use disorder can have periods of addiction and sobriety, or care partners can experience periods of depression and recovery. Potential censoring of the data requires special handling. Overlaying this are the missing at-risk periods for the primary event type when individuals have initiated the primary event but not reached the subsequent secondary event. In this paper, we develop a framework for regression analysis of censored alternating recurrent events that uses a random forest inverse probability weighting strategy to avoid bias in the analysis of the time to the primary event due to informative missingness from the alternate secondary state. The proposed regression model estimates $τ$-restricted mean time to the primary event of interest while taking into account complexities of censored. Simulations show good performance of our method when the alternate times-to-event are either independent or correlated. We analyze a mobile health study data to evaluate the impact of self-care push notifications on the mental state of caregivers of traumatic brain injury patients.

stat.ME↗

arXiv2025

PyCFRL: A Python library for counterfactually fair offline reinforcement learning via sequential data preprocessing

Reinforcement learning (RL) aims to learn and evaluate a sequential decision rule, often referred to as a "policy", that maximizes the population-level benefit in an environment across possibly infinitely many time steps. However, the sequential decisions made by an RL algorithm, while optimized to maximize overall population benefits, may disadvantage certain individuals who are in minority or socioeconomically disadvantaged groups. To address this problem, we introduce PyCFRL, a Python library for ensuring counterfactual fairness in offline RL. PyCFRL implements a novel data preprocessing algorithm for learning counterfactually fair RL policies from offline datasets and provides tools to evaluate the values and counterfactual unfairness levels of RL policies. We describe the high-level functionalities of PyCFRL and demonstrate one of its major use cases through a data example. The library is publicly available on PyPI and Github (https://github.com/JianhanZhang/PyCFRL), and detailed tutorials can be found in the PyCFRL documentation (https://pycfrl-documentation.netlify.app).

stat.ML↗

arXiv2025

Generalized Fitted Q-Iteration with Clustered Data

This paper focuses on reinforcement learning (RL) with clustered data, which is commonly encountered in healthcare applications. We propose a generalized fitted Q-iteration (FQI) algorithm that incorporates generalized estimating equations into policy learning to handle the intra-cluster correlations. Theoretically, we demonstrate (i) the optimalities of our Q-function and policy estimators when the correlation structure is correctly specified, and (ii) their consistencies when the structure is mis-specified. Empirically, through simulations and analyses of a mobile health dataset, we find the proposed generalized FQI achieves, on average, a half reduction in regret compared to the standard FQI.

cs.LG↗

arXiv2025

Generating Synthetic Electronic Health Record Data: a Methodological Scoping Review with Benchmarking on Phenotype Data and Open-Source Software

We conduct a scoping review of existing approaches for synthetic EHR data generation, and benchmark major methods with proposed open-source software to offer recommendations for practitioners. We search three academic databases for our scoping review. Methods are benchmarked on open-source EHR datasets, MIMIC-III/IV. Seven existing methods covering major categories and two baseline methods are implemented and compared. Evaluation metrics concern data fidelity, downstream utility, privacy protection, and computational cost. 42 studies are identified and classified into five categories. Seven open-source methods covering all categories are selected, trained on MIMIC-III, and evaluated on MIMIC-III or MIMIC-IV for transportability considerations. Among them, GAN-based methods demonstrate competitive performance in fidelity and utility on MIMIC-III; rule-based methods excel in privacy protection. Similar findings are observed on MIMIC-IV, except that GAN-based methods further outperform the baseline methods in preserving fidelity. A Python package, "SynthEHRella", is provided to integrate various choices of approaches and evaluation metrics, enabling more streamlined exploration and evaluation of multiple methods. We found that method choice is governed by the relative importance of the evaluation metrics in downstream use cases. We provide a decision tree to guide the choice among the benchmarked methods. Based on the decision tree, GAN-based methods excel when distributional shifts exist between the training and testing populations. Otherwise, CorGAN and MedGAN are most suitable for association modeling and predictive modeling, respectively. Future research should prioritize enhancing fidelity of the synthetic data while controlling privacy exposure, and comprehensive benchmarking of longitudinal or conditional generation methods.

cs.LG↗

arXiv2025

Geometry-driven Bayesian Inference for Ultrametric Covariance Matrices

Ultrametric matrices are a class of covariance matrices that arise in latent tree models. As a parameter space in a statistical model, the set of ultrametric matrices is neither convex nor a smooth manifold. Focus in the literature has hitherto been restricted to estimation through projections and relaxation-based techniques, and inferential methods are lacking. Motivated by this, we establish a bijection between the set of positive definite ultrametric matrices and the set of rooted, leaf-labeled trees equipped with the stratified geometry of the well-known phylogenetic treespace. Using the pullback geometry under the bijection and by adapting sampling algorithms in Bayesian phylogenetics, we develop algorithms to sample from the posterior distribution on the set of ultrametric matrices in a Bayesian latent tree model where the tree may be binary or multifurcating. We demonstrate the utility of the algorithms in simulation studies, and illustrate them on a pre-clinical cancer application to quantify uncertainty about treatment trees that identify treatments with high mechanism similarity that target correlated pathways.

stat.ME↗

arXiv2025

Tree-Regularized Bayesian Latent Class Analysis for Improving Weakly Separated Dietary Pattern Subtyping in Small-Sized Subpopulations

Dietary patterns synthesize multiple related diet components, which can be used by nutrition researchers to examine diet-disease relationships. Latent class models (LCMs) have been used to derive dietary patterns from dietary intake assessment, where each class profile represents the probabilities of exposure to a set of diet components. However, LCM-derived dietary patterns can exhibit strong similarities, or weak separation, resulting in numerical and inferential instabilities that challenge scientific interpretation. This issue is exacerbated in small-sized subpopulations. To address these issues, we provide a simple solution that empowers LCMs to improve dietary pattern estimation. We develop a tree-regularized Bayesian LCM that shares statistical strength between dietary patterns to make better estimates using limited data. This is achieved via a Dirichlet diffusion tree process that specifies a prior distribution for the unknown tree over classes. Dietary patterns that share proximity to one another in the tree are shrunk towards ancestral dietary patterns a priori, with the degree of shrinkage varying across pre-specified food groups. Using dietary intake data from the Hispanic Community Health Study/Study of Latinos, we apply the proposed approach to a sample of 496 US adults of South American ethnic background to identify and compare dietary patterns.

stat.ME↗

arXiv2025

Joint Modeling of Multiple Longitudinal Biomarkers and Survival Outcomes via Threshold Regression: Variability as a Predictor

Longitudinal biomarker data and health outcomes are routinely collected in many studies to assess how biomarker trajectories predict health outcomes. Existing methods primarily focus on mean biomarker profiles, treating variability as a nuisance. However, excess variability may indicate system dysregulations that may be associated with poor outcomes. In this paper, we address the long-standing problem of using variability information of multiple longitudinal biomarkers in time-to-event analyses by formulating and studying a Bayesian joint model. We first model multiple longitudinal biomarkers, some of which are subject to limit-of-detection censoring. We then model the survival times by incorporating random effects and variances from the longitudinal component as predictors through threshold regression that admits non-proportional hazards. We demonstrate the operating characteristics of the proposed joint model through simulations and apply it to data from the Study of Women's Health Across the Nation (SWAN) to investigate the impact of the mean and variability of follicle-stimulating hormone (FSH) and anti-Mullerian hormone (AMH) on age at the final menstrual period (FMP).

stat.AP↗

arXiv2025

Incorporating Auxiliary Variables to Improve the Efficiency of Time-Varying Treatment Effect Estimation

Contextual sensing and delivery of digital interventions to improve health outcomes have gained significant traction in behavioral and psychiatric studies. Micro-randomized trials (MRTs) are a common experimental design for obtaining data-driven evidence on the effectiveness of digital interventions where each individual is repeatedly randomized to receive treatments over numerous time points. Throughout the study, individual characteristics and contextual factors around randomization are collected, with some prespecified as moderators for assessing time-varying causal effect moderation. However, many additional measurements beyond these moderators often go underutilized. Some of these may influence treatment randomization or known to strongly moderate the treatment effect. Incorporating such auxiliary information into the estimation procedure can reduce chance imbalances and improve asymptotic estimation efficiency. In this work, we propose a method to adjust for auxiliary variables in consistently estimating time-varying intervention effects. The approach can also be extended to include post-treatment auxiliary variables when evaluating lagged treatment effects. Under specific conditions, local efficiency gains are guaranteed. We demonstrate the method's utility through simulation studies and an analysis of data from the Intern Health Study (NeCamp et al., 2020).

stat.ME↗

arXiv2025

Doubly Inhomogeneous Reinforcement Learning

This paper studies reinforcement learning (RL) in doubly inhomogeneous environments under temporal non-stationarity and subject heterogeneity. In a number of applications, it is commonplace to encounter datasets generated by system dynamics that may change over time and population, challenging high-quality sequential decision making. Nonetheless, most existing RL solutions require either temporal stationarity or subject homogeneity, which would result in sub-optimal policies if both assumptions were violated. To address both challenges simultaneously, we propose an original algorithm to determine the ``best data chunks" that display similar dynamics over time and across individuals for policy learning, which alternates between most recent change point detection and cluster identification. Our method is general, and works with a wide range of clustering and change point detection algorithms. It is multiply robust in the sense that it takes multiple initial estimators as input and only requires one of them to be consistent. Moreover, by borrowing information over time and population, it allows us to detect weaker signals and has better convergence properties when compared to applying the clustering algorithm per time or the change point detection algorithm per subject. Empirically, we demonstrate the usefulness of our method through extensive simulations and a real data application.

stat.ML↗