A5000822072- Statistical Methods and Inference
- Advanced Statistical Methods and Models
- Statistical Methods and Bayesian Inference
- Nutritional Studies and Diet
- Bayesian Methods and Mixture Models
- Advanced Statistical Process Monitoring
- Statistical Methods in Clinical Trials
- Advanced Causal Inference Techniques
- Gene expression and cancer classification
- Optimal Experimental Design Methods
- Obesity, Physical Activity, Diet
- Genetic Associations and Epidemiology
- Genetic and phenotypic traits in livestock
- Pancreatic function and diabetes
- Statistical and numerical algorithms
- Gaussian Processes and Bayesian Inference
- Control Systems and Identification
- Spectroscopy and Chemometric Analyses
- Neural Networks and Applications
- Consumer Attitudes and Food Labeling
- Diet and metabolism studies
- Bioinformatics and Genomic Networks
- Spatial and Panel Data Analysis
- Genetic Mapping and Diversity in Plants and Animals
- Soil Geostatistics and Mapping
Texas A&M University
2016-2025
Wake Forest University
2011-2025
University of Technology Sydney
2014-2024
Mitchell Institute
2006-2024
College Station Medical Center
2008-2023
Park University
2021-2023
Université Paris Cité
2021-2023
Walter de Gruyter (Germany)
2021-2023
Australian Mathematical Sciences Institute
2022
Renmin University of China
2017-2019
Introduction. Generalized Least Squares and the Analysis of Heteroscedasticity. Estimation Inference for Variance Functions. The Transform-Both-Sides Methodology. Combining Transformations Weighting. Influence Robustness. Technical Complements. Some Open Problems. References. Index.
Abstract The typical generalized linear model for a regression of response Y on predictors (X, Z) has conditional mean function based combination Z). We generalize these models to have nonparametric component, replacing the α T 0X + β 0Z by η0(α 0X) 0Z, where η0(·) is an unknown function. call partially single-index (GPLSIM). include "single-index" models, which β0 = 0. Using local methods, we propose estimates parameters (α0, β0) and obtain their asymptotic distributions. Examples...
Abstract Heteroscedastic regression models are used in fields including economics, engineering, and the biological physical sciences. Often, heteroscedasticity is modeled as a function of covariates or other structural parameters. Standard asymptotic theory implies that how one estimates variance function, particular parameters, has no effect on first-order properties parameter estimates; there evidence, however, both practice higher-order to suggest does matter. Further, some settings,...
The sandwich estimator, also known as robust covariance matrix heteroscedasticity-consistent estimate, or empirical has achieved increasing use in the econometric literature well with growing popularity of generalized estimating equations. Its virtue is that it provides consistent estimates for parameter even when fitted parametric model fails to hold not specified. Surprisingly though, there been little discussion properties method other than consistency. We investigate estimator...
Abstract Suppose that the sum of two independent random variables X and Z is observed, where denotes measurement error has a known distribution, unknown density f to be estimated. One application estimation prior for sequence location parameters. A second arises in errors-in-variables problem nonlinear generalized linear models, when one attempts model distribution true but unobservable covariates. This article shows if normally distributed k bounded derivatives, then fastest attainable...
Abstract It is of interest to estimate the distribution usual nutrient intake for a population from repeat 24‐h dietary recall assessments. A mixed effects model and quantile estimation procedure, developed at National Cancer Institute (NCI), may be used this purpose. The incorporates Box–Cox parameter covariates daily nutrients; parameters are estimated via quasi‐Newton optimization likelihood approximated by adaptive Gaussian quadrature. estimates in Monte Carlo approach generate empirical...
Abstract. We reviewed all studies (since 1945) reporting recurrence rates for treatment of recurrent (previously treated) basal cell carcinomas (BCC) using surgical excision, radiotherapy, cryotherapy, curettage and electrodesiccation, Mohs micrographic surgery. The 5‐year rate surgery is 5.6%. non‐Mohs modalities 19.9% nearly four times higher. Individual the are 17.4% 40.0% 9.8% radiation therapy. There no data cryotherapy. However, 13.0% cryotherapy when follow‐up period less than five...
Abstract We consider two methods of defining a regression analog to trimmed mean. The first was suggested by Koenker and Bassett uses their concept quantiles. Its asymptotic behavior is completely analogous that second method residuals from preliminary estimator. depends heavily on the estimate; it behaves, in general, quite differently than estimator proposed Bassett, can be inefficient at normal model even if percentage trimming small. However, average quantiles used with Bassett's...
The prohormone convertase SPC2 (PC2) participates in the processing of proinsulin, proglucagon, and a variety other neuroendocrine precursors, acting either alone or conjunction with structurally related dense-core granule SPC3 (PC3/PC1). We have generated strain mice lacking active by introducing neomycin resistance gene (Neo r ) into third exon mSPC2 gene. This insertion results synthesis an 3-deleted form that does not undergo autoactivation is secreted. homozygous mutant appear to be...
This paper considers estimation of a continuous bounded probability density when observations from the are contaminated by additive measurement errors having known distribution. Properties estimator obtained deconvolving kernel observed data investigated. When used is sufficiently smooth deconvolved shown to be pointwise consistent and bounds on its integrated mean squared error derived. Very weak assumptions made measurement-error thereby permitting comparison effects different types
Background Most large cohort studies have used a food frequency questionnaire (FFQ) for assessing dietary intake. Several biomarker studies, however, cast doubt on whether the FFQ has sufficient precision to allow detection of moderate but important diet–disease associations. We use data from Observing Protein and Energy Nutrition (OPEN) study compare performance with that 24-hour recall (24HR).
Increasingly, scientific studies yield functional data, in which the ideal units of observation are curves and observed data consist sets that sampled on a fine grid. We present new methodology generalizes linear mixed model to framework, with fitting done by using Bayesian wavelet-based approach. This method is flexible, allowing functions arbitrary form full range fixed effects structures between-curve covariance available framework. It yields nonparametric estimates random-effects as well...