arXiv subjects

Matthew D. Koslovsky

Publications and source records attributed to Matthew D. Koslovsky.

10 recordsLinked to original sources

arXiv2026

A Comprehensive Bayesian Approach to Entity Resolution for Data with Multiple Truths

In many applications, from government to ecology, integrating data from diverse and noisy sources is critical for downstream inference. However, a unique identifier to link records cleanly from the same entity may not exist. Entity resolution (also referred to as de-duplication or record linkage) merges such databases to identify duplicates, allowing for more complete data for inference. A multitude of methods from statistics and computer science in recent years assume a single, immutable true value for each variable used in linkage. We argue this restrictive assumption is often violated in practice, leading researchers to discard useful data and biasing downstream inference. For example, a respondent's education status may truly change between two different surveys, yet existing methods would treat at least one observation as a distorted version of the truth, even though both are correct. In this paper, we propose a novel entity resolution model that comprehensively accommodates "multiple truths" by introducing a mixture of exponential family distributions that further handles multiple data types. We provide options to fit the model with Markov chain Monte Carlo routines and variational inference for massive datasets. We demonstrate the method's value via simulation and linking a longitudinal survey of Italian household wealth.

stat.ME↗

arXiv2026

A Bayesian Functional Concurrent Zero-Inflated Dirichlet-Multinomial Regression Model with Application to Infant Microbiome

The infant microbiome undergoes rapid changes in composition over time and is associated with long-term risks of conditions such as immune strength, allergy, asthma, and other health outcomes. Modeling the associations between exposures or treatments and microbial composition over time is essential for understanding the factors that drive these changes. Estimating these temporal dynamics has several challenges including: repeated measures, overdispersion, compositionality, high-dimensional parameter spaces, and zero-inflation. Many longitudinal regression models used in human microbiome research assume constant effects over time that cannot capture time-varying or functional effects of exposures, ignore the compositional structure of the data by modeling each taxon separately, and are not equipped to handle potential zero-inflation. Dirichlet-multinomial (DM) regression models inherently accommodate overdispersion and the compositional structure of the data and have been extended to account for excess zeros. However, existing DM-based regression models are unable to additionally handle repeated measures designs. To fill this gap, we propose a functional concurrent zero-inflated Dirichlet-multinomial (FunC-ZIDM) regression model which is designed to model time-varying relations between observed covariates and microbial taxa while accounting for zero-inflation, compositionality, and repeated measures. Through simulation, we demonstrate that the model can accurately estimate the underlying functional relations and scale to large compositional spaces. We apply our model to investigate time-varying associations between infant microbiome composition and observed covariates during the 11-week postnatal period. We found that $α$-diversity (i.e., diversity of the microbiome within an individual) is positively associated with a higher gestational age and percentage of breast milk in the diet.

stat.ME↗

arXiv2025

A Bayesian Semiparametric Mixture Model for Clustering Zero-Inflated Microbiome Data

Microbiome research has immense potential for unlocking insights into human health and disease. A common goal in human microbiome research is identifying subgroups of individuals with similar microbial composition that may be linked to specific health states or environmental exposures. However, existing clustering methods are often not equipped to accommodate the complex structure of microbiome data and typically make limiting assumptions regarding the number of clusters in the data which can bias inference. Designed for zero-inflated multivariate compositional count data collected in microbiome research, we propose a novel Bayesian semiparametric mixture modeling framework that simultaneously learns the number of clusters in the data while performing cluster allocation. In simulation, we demonstrate the clustering performance of our method compared to distance- and model-based alternatives and the importance of accommodating zero-inflation when present in the data. We then apply the model to identify clusters in microbiome data collected in a study designed to investigate the relation between gut microbial composition and enteric diarrheal disease.

stat.ME↗

arXiv2024

A Unified Bayesian Framework for Modeling Measurement Error in Multinomial Data

Measurement error in multinomial data is a well-known and well-studied inferential problem that is encountered in many fields, including engineering, biomedical and omics research, ecology, finance, official statistics, and social sciences. Methods developed to accommodate measurement error in multinomial data are typically equipped to handle false negatives or false positives, but not both. We provide a unified framework for accommodating both forms of measurement error using a Bayesian hierarchical approach. We demonstrate the proposed method's performance on simulated data and apply it to acoustic bat monitoring and official crime data.

stat.ME↗

arXiv2024

A Bayesian Nonparametric Approach for Clustering Functional Trajectories over Time

Functional concurrent, or varying-coefficient, regression models are commonly used in biomedical and clinical settings to investigate how the relation between an outcome and observed covariate varies as a function of another covariate. In this work, we propose a Bayesian nonparametric approach to investigate how clusters of these functional relations evolve over time. Our model clusters individual functional trajectories within and across time periods while flexibly accommodating the evolution of the partitions across time periods with covariates. Motivated by mobile health data collected in a novel, smartphone-based smoking cessation intervention study, we demonstrate how our proposed method can simultaneously cluster functional trajectories, accommodate temporal dependence, and provide insights into the transitions between functional clusters over time.

stat.ME↗

arXiv2023

A Bayesian Joint Model for Compositional Mediation Effect Selection in Microbiome Data

Analyzing multivariate count data generated by high-throughput sequencing technology in microbiome research studies is challenging due to the high-dimensional and compositional structure of the data and overdispersion. In practice, researchers are often interested in investigating how the microbiome may mediate the relation between an assigned treatment and an observed phenotypic response. Existing approaches designed for compositional mediation analysis are unable to simultaneously determine the presence of direct effects, relative indirect effects, and overall indirect effects, while quantifying their uncertainty. We propose a formulation of a Bayesian joint model for compositional data that allows for the identification, estimation, and uncertainty quantification of various causal estimands in high-dimensional mediation analysis. We conduct simulation studies and compare our method's mediation effects selection performance with existing methods. Finally, we apply our method to a benchmark data set investigating the sub-therapeutic antibiotic treatment effect on body weight in early-life mice.

stat.AP↗

arXiv2023

A Bayesian Zero-Inflated Dirichlet-Multinomial Regression Model for Multivariate Compositional Count Data

The Dirichlet-multinomial (DM) distribution plays a fundamental role in modern statistical methodology development and application. Recently, the DM distribution and its variants have been used extensively to model multivariate count data generated by high-throughput sequencing technology in omics research due to its ability to accommodate the compositional structure of the data as well as overdispersion. A major limitation of the DM distribution is that it is unable to handle excess zeros typically found in practice which may bias inference. To fill this gap, we propose a novel Bayesian zero-inflated DM model for multivariate compositional count data with excess zeros. We then extend our approach to regression settings and embed sparsity-inducing priors to perform variable selection for high-dimensional covariate spaces. Throughout, modeling decisions are made to boost scalability without sacrificing interpretability or imposing limiting assumptions. Extensive simulations and an application to a human gut microbiome data set are presented to compare the performance of the proposed method to existing approaches. We provide an accompanying R package with a user-friendly vignette to apply our method to other data sets.

stat.ME↗

arXiv2022

Infinite Hidden Markov Models for Multiple Multivariate Time Series with Missing Data

Exposure to air pollution is associated with increased morbidity and mortality. Recent technological advancements permit the collection of time-resolved personal exposure data. Such data are often incomplete with missing observations and exposures below the limit of detection, which limit their use in health effects studies. In this paper we develop an infinite hidden Markov model for multiple asynchronous multivariate time series with missing data. Our model is designed to include covariates that can inform transitions among hidden states. We implement beam sampling, a combination of slice sampling and dynamic programming, to sample the hidden states, and a Bayesian multiple imputation algorithm to impute missing data. In simulation studies, our model excels in estimating hidden states and state-specific means and imputing observations that are missing at random or below the limit of detection. We validate our imputation approach on data from the Fort Collins Commuter Study. We show that the estimated hidden states improve imputations for data that are missing at random compared to existing approaches. In a case study of the Fort Collins Commuter Study, we describe the inferential gains obtained from our model including improved imputation of missing data and the ability to identify shared patterns in activity and exposure among repeated sampling days for individuals and among distinct individuals.

stat.ME↗

arXiv2021

A Bayesian model of microbiome data for simultaneous identification of covariate associations and prediction of phenotypic outcomes

One of the major research questions regarding human microbiome studies is the feasibility of designing interventions that modulate the composition of the microbiome to promote health and cure disease. This requires extensive understanding of the modulating factors of the microbiome, such as dietary intake, as well as the relation between microbial composition and phenotypic outcomes, such as body mass index (BMI). Previous efforts have modeled these data separately, employing two-step approaches that can produce biased interpretations of the results. Here, we propose a Bayesian joint model that simultaneously identifies clinical covariates associated with microbial composition data and predicts a phenotypic response using information contained in the compositional data. Using spike-and-slab priors, our approach can handle high-dimensional compositional as well as clinical data. Additionally, we accommodate the compositional structure of the data via balances and overdispersion typically found in microbial samples. We apply our model to understand the relations between dietary intake, microbial samples, and BMI. In this analysis, we find numerous associations between microbial taxa and dietary factors that may lead to a microbiome that is generally more hospitable to the development of chronic diseases, such as obesity. Additionally, we demonstrate on simulated data how our method outperforms two-step approaches and also present a sensitivity analysis.

stat.ME↗

arXiv2020

A Bayesian Time-Varying Effect Model for Behavioral mHealth Data

The integration of mobile health (mHealth) devices into behavioral health research has fundamentally changed the way researchers and interventionalists are able to collect data as well as deploy and evaluate intervention strategies. In these studies, researchers often collect intensive longitudinal data (ILD) using ecological momentary assessment methods, which aim to capture psychological, emotional, and environmental factors that may relate to a behavioral outcome in near real-time. In order to investigate ILD collected in a novel, smartphone-based smoking cessation study, we propose a Bayesian variable selection approach for time-varying effect models, designed to identify dynamic relations between potential risk factors and smoking behaviors in the critical moments around a quit attempt. We use parameter-expansion and data-augmentation techniques to efficiently explore how the underlying structure of these relations varies over time and across subjects. We achieve deeper insights into these relations by introducing nonparametric priors for regression coefficients that cluster similar effects for risk factors while simultaneously determining their inclusion. Results indicate that our approach is well-positioned to help researchers effectively evaluate, design, and deliver tailored intervention strategies in the critical moments surrounding a quit attempt.

stat.ME↗