A5113720094- Statistical Methods and Inference
- Genetic Neurodegenerative Diseases
- Statistical Methods and Bayesian Inference
- Advanced Causal Inference Techniques
- Fibromyalgia and Chronic Fatigue Syndrome Research
- Neurological disorders and treatments
- Statistical Methods in Clinical Trials
- Health disparities and outcomes
- Health Systems, Economic Evaluations, Quality of Life
- Parkinson's Disease Mechanisms and Treatments
- Bayesian Methods and Mixture Models
- Mitochondrial Function and Pathology
- Smoking Behavior and Cessation
- SARS-CoV-2 detection and testing
- Metabolomics and Mass Spectrometry Studies
- Genetic Associations and Epidemiology
- Biochemical Analysis and Sensing Techniques
- Migration, Aging, and Tourism Studies
- Advanced Statistical Methods and Models
- Biosensors and Analytical Detection
- Alcohol Consumption and Health Effects
- Bioinformatics and Genomic Networks
- Genetic Mapping and Diversity in Plants and Animals
- Electoral Systems and Political Participation
- Statistical and Computational Modeling
University of North Carolina at Chapel Hill
2020-2025
University of Puerto Rico at Carolina
2024
University of Mississippi Medical Center
2022
Texas A&M University
2011-2020
RAND Corporation
2019
Texas A&M Health Science Center
2013-2017
Columbia University
2014-2017
Pennsylvania State University
2017
Columbia University Irving Medical Center
2017
The University of Sydney
2013-2016
Modeling symptom progression to identify ideal subjects for a Huntington's disease clinical trial is problematic since time diagnosis, key covariate, can be heavily censored. Imputation an appealing strategy that replaces the censored covariate with its conditional mean, but existing methods saw over 200% bias under heavy censoring. Calculating means well requires estimating and then integrating survival function of from value infinity. To estimate flexibly, use semiparametric Cox model...
Abstract Background and Objectives While Hispanic/Latino populations in the U.S. are remarkably diverse terms of birthplace age at migration, we poorly understand how these factors associated with cognitive aging. Our research seeks to operationalize a life course perspective migration health contribute new understanding Alzheimer’s disease / related dementias among U.S.-based older adults. Research Design Methods Harnessing Hispanic Community Health Study/Study Latinos (n=16,415) Study...
Objective: Spiritual well-being (SWB) has been shown to delay the onset of cognitive decline among older adults predisposed Alzheimer's disease and related neurodegenerative dementias. It was, however, unknown if SWB is also associated with in manifestation ("phenoconversion") rare, genetic dementias, such as Huntington's (HD). Thus, we sought evaluate association between phenocovnersion people at-risk for HD. Methods: The "Prospective Huntington At Risk Observation Study" (PHAROS), a...
Abstract Motivation: Gut microbiota can be classified at multiple taxonomy levels. Strategies to use changes in composition effect health improvements require knowing which level interventions should aimed. Identifying these important levels is difficult, however, because most statistical methods only consider when the are one level, not multiple. Results: Using L1 and L2 regularizations, we developed a new variable selection method that identifies features The regularization parameters...
Background and AimsTo investigate associations between avocado intake glycemia in adults with Hispanic/Latino ancestry.Methods ResultsThe of measures insulin glucose homeostasis were evaluated a cross-sectional analysis up to 14,591 adults, using of: average levels (hemoglobin A1c; HbA1c), fasting insulin, after an oral tolerance test (OGTT), calculated resistance (HOMA-IR, HOMA-%β), insulinogenic index. Associations assessed multivariable linear regression models, which controlled for...
In certain genetic studies, clinicians and counselors are interested in estimating the cumulative risk of a disease for individuals with without rare deleterious mutation. Estimating is difficult, however, when estimates based on family history data. Often, mutation status many members unknown; instead, only estimated probabilities patient having available. Also, ages disease-onset subject to right censoring. Existing methods estimate using such family-based data provide estimation at...
Abstract Introduction We studied the replication and generalization of previously identified metabolites potentially associated with global cognitive function in multiple race/ethnicities assessed contribution diet to these associations. Methods tested metabolite‐cognitive associations U.S.A. Hispanic/Latino adults ( n = 2222) from Community Health Study/ Study Latinos (HCHS/SOL) European 1365) African 478) Americans Atherosclerosis Risk In Communities (ARIC) Study. applied Mendelian...
This work presents methods for estimating genotype-specific distributions from genetic epidemiology studies where the event times are subject to right censoring, genotypes not directly observed, and data arise a mixture of scientifically meaningful subpopulations. Examples such include kin-cohort quantitative trait locus (QTL) studies. Current analyzing censored two types nonparametric maximum likelihood estimators (NPMLEs) which do make parametric assumptions on density functions. Although...
Amino acid nutrition studies often involve repeated measures data. An example is that the concentrations of plasma citrulline in steers are repeatedly measured from same animals. The standard ANOVA method does not detect significant time changes within 6 hours after consumed rumen-protected citrulline, while a graphical analysis indicates there exists effect. Here we describe three mixed model analyses capture effect statistically way, accounting for correlations measurements over steers....
Traditional crash count models, such as the Poisson and negative binomial do not account for temporal correlation of data. In reality, crashes that occur in same time frame are likely to share unobserved effects may have been excluded from model. If data is ignored, estimated parameters can be biased less precise. Therefore, there a need extend standard models by incorporating dependence. Whereas literature modeling series well developed, its applications traffic limited. A particularly...
Mega-analysis, or the meta-analysis of individual data, enables pooling and comparing multiple studies to enhance estimation power. A challenge in mega-analysis is estimating distribution for clustered, potentially censored event times where dependency structure can introduce bias if ignored. We propose a new proportional odds model with unknown, time-varying coefficients, random effects. The directly captures dependencies, handles censoring using pseudo-values, permits simple by...
Logistic models with a random intercept are prevalent in medical and social research where clustered longitudinal data often collected. Traditionally, the these is assumed to follow some parametric distribution such as normal distribution. However, an assumption inevitably raises concerns about model misspecification misleading inference conclusions, especially when there dependence between covariates. To protect against issues, we use semiparametric approach develop computationally simple...
The landscape of survival analysis is constantly being revolutionized to answer biomedical challenges, most recently the statistical challenge censored covariates rather than outcomes. There are many promising strategies tackle covariates, including weighting, imputation, maximum likelihood, and Bayesian methods. Still, this a relatively fresh area research, different from areas outcomes (i.e., analysis) or missing covariates. In review, we discuss unique challenges encountered when handling...
Longitudinal studies are prevalent in biological and social sciences where subjects measured repeatedly over time. Modeling the correlations handling missing data among most challenging problems analyzing such data. There various methods for data, but data-based graphical modeling covariance matrix of longitudinal relatively new. We adopt an approach based on modified Cholesky decomposition which handles both challenges. It amounts to formulating parametric models regression coefficients...
Biomedical studies of neuroimaging and genomics collect large amounts data on a small subset subjects so as to not miss informative predictors. An important goal is identifying those predictors that provide better visualization the could serve cost-effective measures for future clinical trials. Identifying such challenging, however, when are naturally interrelated response failure time prone censoring. We propose handle these challenges with novel variable selection technique. Our approach...
An important goal in clinical and statistical research is properly modeling the distribution for clustered failure times which have a natural intra-class dependency are subject to censoring. We handle these challenges with novel approach that does not impose restrictive or distributional assumptions. Using logit transformation, we relate covariates random, subject-specific effect. The modeled unknown functional forms, random effect may depend on an unspecified distribution. introduce...
Background:Critical to discovering targeted therapies for Huntington disease (HD) are validated methods that more precisely predict when clinical outcomes occur different patient profiles. Objective:To the probability of motor diagnosis (diagnostic confidence level 4 ) on Unified Huntington's Disease Rating Scale (UHDRS), cognitive impairment (two or neuropsychological scores UHDRS were 1.5 standard deviations below normative means) and Stage II Total Functional Capacity (TFC) first by...