Statistics and Probability Commons^™

Bayesian Change-Point Analysis In Linear Regression Model With Scale Mixtures Of Normal Distributions, Shuaimin Kang

Dissertations, Master's Theses and Master's Reports - Open

In this thesis, we consider Bayesian inference on the detection of variance change-point models with scale mixtures of normal (for short SMN) distributions. This class of distributions is symmetric and thick-tailed and includes as special cases: Gaussian, Student-t, contaminated normal, and slash distributions. The proposed models provide greater flexibility to analyze a lot of practical data, which often show heavy-tail and may not satisfy the normal assumption.

As to the Bayesian analysis, we specify some prior distributions for the unknown parameters in the variance change-point models with the SMN distributions. Due to the complexity of the joint posterior …

Variable Selection With False Discovery Control, Kevin He, Yanming Li, Ji Zhu, Hongliang Liu, Jeffrey E. Lee, Christopher I. Amos, Terry Hyslop, Jiashun Jin, Qinyi Wei, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

Technological advances that allow routine identification of high-dimensional risk factors have led to high demand for statistical techniques that enable full utilization of these rich sources of information for genome-wide association studies (GWAS). Variable selection for censored outcome data as well as control of false discoveries (i.e. inclusion of irrelevant variables) in the presence of high-dimensional predictors present serious challenges. In the context of survival analysis with high-dimensional covariates, this paper develops a computationally feasible method for building general risk prediction models, while controlling false discoveries. We have proposed a high-dimensional variable selection method by incorporating stability selection to control …

Tests For Gene-Environment Interactions And Joint Effects With Exposure Misclassification, Philip S. Boonstra, Bhramar Mukherjee, Stephen B. Gruber, Jaeil Ahn, Stephanie L. Schmit, Nilanjan Chatterjee

The University of Michigan Department of Biostatistics Working Paper Series

The number of methods for genome-wide testing of gene-environment interactions (GEI) continues to increase with the hope of discovering new genetic risk factors and obtaining insight into the disease-gene-environment relationship. The relative performance of these methods based on family-wise type 1 error rate and power depends on underlying disease-gene-environment associations, estimates of which may be biased in the presence of exposure misclassification. This simulation study expands on a previously published simulation study of methods for detecting GEI by evaluating the impact of exposure misclassification. We consider seven single step and modular screening methods for identifying GEI at a genome-wide level …

Voc Emissions From Beef Feedlot Pen Surfaces As Affected By Within-Pen Location, Moisture And Temperature, Bryan L. Woodbury, John E. Gilley, David B. Parker, David B. Marx, Roger A. Eigenberg

Department of Statistics: Faculty Publications

A laboratory study was conducted to evaluate the effects of pen location, moisture, and temperature on emissions of volatile organic compounds (VOC) from surface materials obtained from feedlot pens where beef cattle were fed a diet containing 30% wet distillers grain plus solubles. Surface materials were collected from the feed trough (bunk), drainage, and raised areas (mounds) within three feedlot pens. The surface materials were mixed with water to represent dry, wet, or saturated conditions and then incubated at temperatures of 5, 15, 25 and 35 C. A wind tunnel and gas chromatograph-mass spectrometer were used to collect and quantify …

The Support Vector Machine And Mixed Integer Linear Programming: Ramp Loss Svm With L1-Norm Regularization, Eric J. Hess, J. Paul Brooks

Statistical Sciences and Operations Research Publications

The support vector machine (SVM) is a flexible classification method that accommodates a kernel trick to learn nonlinear decision rules. The traditional formulation as an optimization problem is a quadratic program. In efforts to reduce computational complexity, some have proposed using an L1-norm regularization to create a linear program (LP). In other efforts aimed at increasing the robustness to outliers, investigators have proposed using the ramp loss which results in what may be expressed as a quadratic integer programming problem (QIP). In this paper, we consider combining these ideas for ramp loss SVM with L1-norm regularization. The result is four …

Firing Rate Dynamics In Recurrent Spiking Neural Networks With Intrinsic And Network Heterogeneity, Cheng Ly

Statistical Sciences and Operations Research Publications

Heterogeneity of neural attributes has recently gained a lot of attention and is increasing recognized as a crucial feature in neural processing. Despite its importance, this physiological feature has traditionally been neglected in theoretical studies of cortical neural networks. Thus, there is still a lot unknown about the consequences of cellular and circuit heterogeneity in spiking neural networks. In particular, combining network or synaptic heterogeneity and intrinsic heterogeneity has yet to be considered systematically despite the fact that both are known to exist and likely have significant roles in neural network dynamics. In a canonical recurrent spiking neural network model, …

Ua56/1 Fact Book, Wku Institutional Research

WKU Administration Documents

Statistical and demographic profile of WKU.

A History Of Undergraduate Education For Public Health: From Behind The Scenes To Center Stage., Richard Kenneth Riegelman, Susan Albertine, Randy Wykoff

GW Biostatistics Center

No abstract provided.

Scaling Down To Scale Up: A Health Economic Analysis Of Integrating Point-Of-Care Syphilis Testing Into Antenatal Care In Zambia During Pilot And National Rollout Implementation., Katharine D. Shelley, Éimhín M. Ansbro, Alexander Tshaka Ncube, Sedona Sweeney, Colette Fleischer, Grace Tembo Mumba, Michelle M. Gill, Susan Strasser, Rosanna W. Peeling, Fern Terris-Prestholt

GW Biostatistics Center

Maternal syphilis results in an estimated 500,000 stillbirths and neonatal deaths annually in Sub-Saharan Africa. Despite the existence of national guidelines for antenatal syphilis screening, syphilis testing is often limited by inadequate laboratory and staff services. Recent availability of inexpensive rapid point-of-care syphilis tests (RST) can improve access to antenatal syphilis screening. A 2010 pilot in Zambia explored the feasibility of integrating RST within prevention of mother-to-child-transmission of HIV services. Following successful demonstration, the Zambian Ministry of Health adopted RSTs into national policy in 2011. Cost data from the pilot and 2012 preliminary national rollout were extracted from project records, …

Treatment-Induced Changes In Plasma Adiponectin Do Not Reduce Urinary Albumin Excretion In The Diabetes Prevention Program Cohort., Kieren J. Mather, Qing Pan, William C. Knowler, Tohru Funahashi, George A. Bray, Et Al.

GW Biostatistics Center

BACKGROUND AND OBJECTIVES: Molecular data suggests that adiponectin may directly regulate urinary albumin excretion. In the Diabetes Prevention Program (DPP) we measured adiponectin and albuminuria before and after intervention, and we previously reported increases in adiponectin with interventions. Here we have used the DPP dataset to test the hypothesis that treatment-related increases in adiponectin may reduce albuminuria in obesity.

DESIGN, SETTING, PARTICIPANTS AND METHODS: We evaluated cross-sectional correlations between plasma adiponectin and urinary albumin excretion at baseline, and the relationship of treatment-related changes in adiponectin and albuminuria. Baseline and follow-up urine albumin to creatinine ratios (ACR (albumin to creatinine ratio)) …

The Cost-Effectiveness Of Diabetes Prevention: Results From The Diabetes Prevention Program And The Diabetes Prevention Program Outcomes Study, William H. Herman

GW Biostatistics Center

Background

The Diabetes Prevention Program (DPP) was a randomized, controlled clinical trial. It demonstrated that among high-risk individuals with impaired glucose tolerance, diabetes incidence was reduced by 58 % with lifestyle intervention and 31 % with metformin compared to placebo. During the Diabetes Prevention Program Outcomes Study (DPPOS), all DPP participants were unmasked to their treatment assignments, the original lifestyle intervention group was offered additional lifestyle support, the metformin group continued metformin, and all three groups were offered a group-implemented lifestyle intervention. Over the 10 years of combined DPP/DPPOS follow-up, diabetes incidence was reduced by 34 % in the lifestyle …

Skin Collagen Advanced Glycation Endproducts (Ages) And The Long-Term Progression Of Sub-Clinical Cardiovascular Disease In Type 1 Diabetes, Vincent M. Monnier, Wanjie Sun, Xiaoyu Gao, David R. Sell, Patricia Cleary, John M. Lachin, Saul Genuth, The Dcct/Edic Research Group

GW Biostatistics Center

BACKGROUND:

We recently reported strong associations between eight skin collagen AGEs and two solubility markers from skin biopsies obtained at DCCT study closeout and the long-term progression of microvascular disease in EDIC, despite adjustment for mean glycemia. Herein we investigated the hypothesis that some of these AGEs (fluorescence to be reported elsewhere) correlate with long-term subclinical cardiovascular disease (CVD) measurements, i.e. coronary artery calcium score (CAC) at EDIC year 7-9 (n = 187), change of carotid intima-media thickness (IMT) from EDIC year 1 to year 6 and 12 (n = 127), and cardiac MRI outcomes at EDIC year 15-16 (n …

Quality Control Measures Over 30 Years In A Multicenter Clinical Study: Results From The Diabetes Control And Complications Trial / Epidemiology Of Diabetes Interventions And Complications (Dcct/Edic) Study., Gayle M Lorenzi, Barbara H Braffett, Valerie L Arends, Ronald P Danis, Lisa Diminick, Kandace A Klumpp, Anthony D Morrison, Elsayed Z Soliman, Michael W Steffes, Patricia A Cleary

GW Biostatistics Center

No abstract provided.

The Cost-Effectiveness Of Diabetes Prevention: Results From The Diabetes Prevention Program And The Diabetes Prevention Program Outcomes Study, William Herman

GW Biostatistics Center

Background

The Diabetes Prevention Program (DPP) was a randomized, controlled clinical trial. It demonstrated that among high-risk individuals with impaired glucose tolerance, diabetes incidence was reduced by 58 % with lifestyle intervention and 31 % with metformin compared to placebo. During the Diabetes Prevention Program Outcomes Study (DPPOS), all DPP participants were unmasked to their treatment assignments, the original lifestyle intervention group was offered additional lifestyle support, the metformin group continued metformin, and all three groups were offered a group-implemented lifestyle intervention. Over the 10 years of combined DPP/DPPOS follow-up, diabetes incidence was reduced by 34 % in the lifestyle …

Pregnant And Postpartum Women’S Experiences And Perspectives On The Acceptability And Feasibility Of Copackaged Medicine For Antenatal Care And Pmtct In Lesotho, Michelle M. Gill, Heather J. Hoffman, Appolinaire Tiam, Florence M. Mohai, Mokone Majoalane, Anthony Isavwa, Laura Guay, +4 Additional Authors

Epidemiology Faculty Publications

Objective. To improve PMTCT and antenatal care-related service delivery, a pack with centrally prepackaged medicine was rolled out to all pregnant women in Lesotho in 2011. This study assessed acceptability and feasibility of this copackaging mechanism for drug delivery among pregnant and postpartum women.

Methods. Acceptability and feasibility were assessed in a mixed method, cross-sectional study through structured interviews (SI) and semistructured interviews (SSI) conducted in 2012 and 2013.

Results. 290 HIV-negative women and 437 HIV-positive women (n=727) participated. Nearly all SI participants found prepackaged medicines acceptable, though modifications such as size reduction of the pack were …

Clinical Significance Of Left Atrial Anatomic Abnormalities Identified By Cardiac Computed Tomography, Ara V. Vehian, Brian G. Choi, Satinder Rekhi, Heather A. Young, Raman S. Dusaj, Robert K. Zeman

Epidemiology Faculty Publications

Purpose: The clinical significance of newly identified left atrial anatomic abnormalities (LAAA)— accessory appendages, diverticula, septal pouches—by multidetector CT (MDCT) remains unclear. Similar anatomical outpouchings, i.e., the left atrial appendage, have been associated with cardioembolisms and arrhythmia. To test the hypothesis that LAAA are also associated with increased risk of these events, we performed a retrospective analysis to examine the association of LAAA in patients undergoing CT with embolic events and arrhythmia.

Methods: 242 patients (mean age 56 SD 12 years, 41% female) were selected who had CT coronary angiography performed with 64-row MDCT between 2007 and 2012 if complete …

Parity And Diabetes Risk Among Hispanic Women From Colombia: Cross-Sectional Evidence, Pablo Cure, Heather J. Hoffman, Carlos Cure-Cure

Epidemiology Faculty Publications

Objective

The association between parity and type 2 diabetes has been studied in developed countries and in Singapore and Chinese women but not in Hispanics. Herein we evaluated the association between parity (number of live births) with diabetes in a group of Hispanic postmenopausal women from Colombia.

Research design and methods

Herein we evaluated the association between parity and diabetes in a population of 1,795 women from Colombia. Women were divided in birth categories (0 [referent], 1 or 2, 3–5, 6 or > births). Medical history of diabetes and anthropometric characteristics were recorded. Logistic regressions were performed in order to find …

Implementation Of Point-Of-Care Diagnostics Leads To Variable Uptake Of Syphilis, Anemia And Cd4+ T-Cell Count Testing In Rural Maternal And Child Health Clinics., Caroline De Schacht, Carlota Lucas, Nádia Sitoe, Rhoderick Machekano, Patrina Chongo, Laura Guay, Et Al.

Epidemiology Faculty Publications

INTRODUCTION: Anemia, syphilis and HIV are high burden diseases among pregnant women in sub-Saharan Africa. A quasi-experimental study was conducted in four health facilities in Southern Mozambique to evaluate the effect of point-of-care technologies for hemoglobin quantification, syphilis testing and CD4+ T-cell enumeration performed within maternal and child health services on testing and treatment coverage, and assessing acceptability by health workers.

METHODS: Demographic and testing data on women attending first antenatal care services were extracted from existing records, before (2011; n = 865) and after (2012; n = 808) introduction of point-of-care testing. Study outcomes per health facility were compared …

A Systematic Review And Meta-Regression Analysis Of Lung Cancer Risk And Inorganic Arsenic In Drinking Water., Steven H. Lamm, Hamid Ferdosi, Elisabeth K. Dissen, Ji Li, Jaeil Ahn

Epidemiology Faculty Publications

High levels (> 200 µg/L) of inorganic arsenic in drinking water are known to be a cause of human lung cancer, but the evidence at lower levels is uncertain. We have sought the epidemiological studies that have examined the dose-response relationship between arsenic levels in drinking water and the risk of lung cancer over a range that includes both high and low levels of arsenic. Regression analysis, based on six studies identified from an electronic search, examined the relationship between the log of the relative risk and the log of the arsenic exposure over a range of 1-1000 µg/L. The …

A Generalized Inflated Geometric Distribution, Ram Datt Joshi

Theses, Dissertations and Capstones

A count data that have excess number of zeros, ones, twos or threes are commonplace in experimental studies. But these inflated frequencies at particular counts may lead to over dispersion and thus may cause difficulty in data analysis. So, to get appropriate results from them and to overcome the possible anomalies in parameter estimation, we may need to consider suitable inflated distribution.

In this thesis, we have considered a Swedish fertility dataset with inflated values at some particular counts. Generally, Inflated Poisson or Inflated Negative Binomial distribution are the most common distributions for analyzing such data. Geometric distribution can be …