A5029794342- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Advanced Causal Inference Techniques
- Advanced Statistical Methods and Models
- Tensor decomposition and applications
- Recommender Systems and Techniques
- Epigenetics and DNA Methylation
- Advanced Neuroimaging Techniques and Applications
- Complex Network Analysis Techniques
- Statistical Methods in Clinical Trials
- Migration, Health and Trauma
- Computational Physics and Python Applications
- Text and Document Classification Technologies
- Gene expression and cancer classification
- Advanced Clustering Algorithms Research
- Spatial and Panel Data Analysis
- Resilience and Mental Health
- Topic Modeling
- Genetic and phenotypic traits in livestock
- Bayesian Modeling and Causal Inference
- Machine Learning and Data Classification
- Advanced Bandit Algorithms Research
- Machine Learning and Algorithms
- Opinion Dynamics and Social Influence
University of California, Irvine
2018-2025
University of Illinois Urbana-Champaign
2011-2021
Seoul National University
2021
McGill University
2017-2021
Decision Sciences (United States)
2020
University of Minnesota
2018-2020
Yale University
2018
Fudan University
2018
Center for Genomic Science
2003-2017
Western Michigan University
2017
Institute of Neurological Disorders and Stroke (NINDS) rt-PA Study Group reported a benefit with intravenous tissue-type plasminogen activator (IV tPA) for patients acute ischemic stroke less than 3 hours' duration. 1The Food Drug Administration (FDA) subsequently approved IV tPA in June 1996.The use has caused dramatic change the way is approached.However, carries 10fold increased risk intracerebral hemorrhage (ICH). 1 To minimize risks thrombolytic treatment, guidelines have been devised...
Journal Article Improving generalised estimating equations using quadratic inference functions Get access Annie Qu, Qu Department of Statistics, Oregon State University, Corvallis, 97331, U.S.A. qu@stat.orst.edu 326 Thomas Building, The Pennsylvania University Park, 16802, bgl@psu.edu bing@stat.psu.edu Search for other works by this author on: Oxford Academic Google Scholar Bruce G. Lindsay, Lindsay Bing Li Biometrika, Volume 87, Issue 4, December 2000, Pages 823–836,...
We consider the penalized generalized estimating equations (GEEs) for analyzing longitudinal data with high-dimensional covariates, which often arise in microarray experiments and large-scale health studies. Existing regression procedures assume independent rely on likelihood function. Construction of a feasible joint function is challenging, particularly correlated discrete outcome data. The GEE procedure only requires specifying first two marginal moments working correlation structure....
Summary Nonparametric smoothing methods are used to model longitudinal data, but the challenge remains incorporate correlation into nonparametric estimation procedures. In this article, we propose an efficient procedure for varying‐coefficient models data. The proposed can easily take account within subjects and deal directly with both continuous discrete response data under framework of generalized linear models. approach yields a more estimator than equation when working is misspecified....
Abstract We consider the generalized additive model when responses from same cluster are correlated. Incorporating correlation in estimation of nonparametric components for is important because it improves efficiency and increases statistical power selection. In our setting, there no specified likelihood function model, outcomes could be nonnormal discrete, which makes selection very challenging problems. propose consistent that incorporate structure. establish an asymptotic property with...
Recommender systems have been widely adopted by electronic commerce and entertainment industries for individualized prediction recommendation, which benefit consumers improve business intelligence. In this article, we propose an innovative method, namely the recommendation engine of multilayers (REM), tensor recommender systems. The proposed method utilizes structure a response to integrate information from multiple modes, creates additional layer nested latent factors accommodate...
For multisource data, blocks of variable information from certain sources are likely missing. Existing methods for handling missing data do not take structures block-wise into consideration. In this article, we propose a multiple imputation (MBI) approach, which incorporates imputations based on both complete and incomplete observations. Specifically, given pattern group, the in MBI incorporate more samples groups with fewer observed variables addition to group We construct estimating...
Summary Model selection for marginal regression analysis of longitudinal data is challenging owing to the presence correlation and difficulty specifying full likelihood, particularly correlated categorical data. The paper introduces a novel Bayesian information criterion type model procedure based on quadratic inference function, which does not require likelihood or quasi-likelihood. With probability approaching 1, selects most parsimonious correct model. Although working matrix assumed,...
Identifying an informative correlation structure is important in improving estimation efficiency for longitudinal data. We approximate the empirical estimator of matrix by groups known basis matrices that represent different structures, and transform selection problem to a covariate problem. To address both complexity informativeness matrix, we minimize objective function consists two parts: difference between information model approximation penalty penalizes models with too many matrices....
Diagnosis of basal-like breast cancer (BLBC) remains a bottleneck to conducting effective clinical trials for this aggressive subtype. We postulated that elevated expression Forkhead Box transcription factor C1 (FOXC1) is simple and accurate diagnostic biomarker BLBC. Accuracy FOXC1 in identifying BLBC was compared with the PAM50 gene panel microarray (GEM) (n = 1992) quantitative real-time polymerase chain reaction (qRT-PCR) 349) datasets. A FOXC1-based immunohistochemical (IHC) assay...
We propose generalized additive partial linear models for complex data which allow one to capture nonlinear patterns of some covariates, in the presence components. The proposed method improves estimation efficiency and increases statistical power correlated through incorporating correlation information. A unique feature is its capability handling model selection cases where it difficult specify likelihood function. derive quadratic inference function-based estimators coefficients...
Because of the accessibility big data collections from consumers, products, and stores, advanced sales forecasting capabilities have drawn great attention many businesses, especially those in retail, because importance decision making. Improvement accuracy, even by a small percentage, may substantial impact on companies’ production financial planning, marketing strategies, inventory controls, supply chain management. Specifically, our research goal is to forecast each product store near...
ABSTRACT Individualized modeling has become increasingly popular in recent years with its growing application fields such as personalized medicine and mobile health studies. With rich longitudinal measurements, it is of great interest to model certain subject‐specific time‐varying covariate effects. In this paper, we propose an individualized nonparametric by leveraging the subgroup information from population. The proposed method approximates effect using B‐splines aggregates estimated...
Background: Sleep disturbances, such as difficulty in falling asleep and multiple awakenings at night, are prevalent among persons with Alzheimer’s disease related dementias (hereafter dementia), resulting advanced cognitive impairment increased behavioral problems. Additionally, family caregivers (eg, spouses or offspring) suffer from reduced sleep quality a result of disturbances the dementia (PWDs) they care for. Relatively little is known about interaction parameters dyads...
Extensive literature has been proposed for the analysis of correlated survival data. Subjects within a cluster share some common characteristics, e.g., genetic and environmental factors, so their time-to-event outcomes are correlated. The frailty model under proportional hazards assumption widely applied clustered outcomes. However, prediction performance this method can be less satisfactory when risk factors have complicated effects, nonlinear interactive. To deal with these issues, we...
Pregnancy is a critical period characterized by profound physiological and psychological adaptations that can significantly impact both maternal fetal health outcomes. Thus, it imperative to implement targeted evidence-based interventions enhance well-being during the prenatal period. Mobile (mHealth) technologies enable continuous, real-time monitoring of states, providing detailed insights into behaviors individual responses in natural settings. This study leveraged mHealth technologies,...
In this article, we propose a heterogeneous modeling framework which achieves individual-wise feature selection and covariates’ effects subgrouping simultaneously. contrast to conventional model approaches, the new approach constructs separation penalty with multidirectional shrinkages, facilitates individualized distinguish strong signals from noisy ones selects different relevant variables for individuals. Meanwhile, proposed identifies subgroups among individuals share similar effects,...
This article provides an overview of tensors, their properties, and applications in statistics. Tensors, also known as multidimensional arrays, are generalizations matrices to higher orders useful data representation architectures. We first review basic tensor concepts decompositions, then we elaborate traditional recent tensors the fields recommender systems imaging analysis. illustrate for network explore relations among interacting units a complex system. Some canonical computational...
Abstract The quadratic inference function (QIF) is a new statistical methodology developed for the estimation and in longitudinal data analysis using marginal models. This method an alternative to popular generalized estimating equations approach, it has several useful properties such as robustness, goodness‐of‐fit test model selection. paper presents introductory review of QIF, with strong emphasis on its applications. In particular, recently SAS MACRO QIF illustrated this obtain numerical...