Statistics and Probability Commons^™

Semiparametric Transformation Models For Semicompeting Survival Data, Huazhen Lin, Ling Zhou, Chunhong Li, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

Semicompeting risk outcome data, e.g. time to disease progression and time to death, are commonly collected in clinical trials, but complicated analytical tools hamper the analysis and the interpretation of the results. We propose a novel semiparametric transformation model for such data. Compared with the existing models, our model is advantageous in the following distinctive ways. First, it allows us to provide direct estimators of the regression analysis and the association parameter. Second, the measure of surrogacy, for example, the proportion of treatment effect and relative effect, can also be directly obtained. We propose a two-stage estimation procedure for inference …

Score Test Variable Screening, Sihai Dave Zhao, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

Variable screening has emerged as a crucial first step in the analysis of high-throughput data, but existing procedures can be computationally cumbersome, difficult to justify theoretically, or inapplicable to certain types of analyses. Motivated by a high-dimensional censored quantile regression problem in multiple myeloma genomics, this paper makes three contributions. First, we establish a score test-based screening framework, which is widely applicable, extremely computationally efficient, and relatively simple to justify. Secondly, we propose a resampling-based procedure for selecting the number of variables to retain after screening according to the principle of reproducibility. Finally, we propose a new iterative score test …

A Latent Variable Transformation Model Approach For Exploring Dysphagia, Anna Snavely, David P. Harrington, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

No abstract provided.

Covariance-Enhanced Discriminant Analysis, Peirong Xu, Ji Zhu, Lixing Zhu, Yi Li

The University of Michigan Department of Biostatistics Working Paper Series

Linear discriminant analysis (LDA), a classical method in pattern recognition and machine learning, has been widely used to characterize or separate multiple classes via linear combinations of features. However, the high-dimensionality of the high-throughput features obtained from modern biological experiments, for example, microarray or proteomics, defies traditional discriminant analysis techniques. The possible interfeature correlations present additional challenges and are often under-utilized in modeling. In this paper, by incorporating the possible inter-feature correlations, we propose a Covariance-Enhanced Discriminant Analysis (CEDA) method that simultaneously and consistently selects informative features and identifies the corresponding discriminable classes. We show that, under mild regularity conditions, …

A Method For Generating Realistic Correlation Matrices, Johanna S. Hardin, Stephan Ramon Garcia, David Golan

Pomona Faculty Publications and Research

Simulating sample correlation matrices is important in many areas of statistics. Approaches such as generating Gaussian data and finding their sample correlation matrix or generating random uniform $[-1,1]$ deviates as pairwise correlations both have drawbacks. We develop an algorithm for adding noise, in a highly controlled manner, to general correlation matrices. In many instances, our method yields results which are superior to those obtained by simply simulating Gaussian data. Moreover, we demonstrate how our general algorithm can be tailored to a number of different correlation models. Using our results with a few different applications, we show that simulating correlation matrices …

Generalized Least-Squares Regressions I: Efficient Derivations, Nataniel Greene

Publications and Research

Ordinary least-squares regression suffers from a fundamental lack of symmetry: the regression line of y given x and the regression line of x given y are not inverses of each other. Alternative symmetric regression methods have been developed to address this concern, notably: orthogonal regression and geometric mean regression. This paper presents in detail a variety of least squares regression methods which may not have been known or fully explicated. The derivation of each method is made efficient through the use of Ehrenberg's formula for the ordinary least-squares error and through the extraction of a weight function g(b) which characterizes …

Information And Communication Technology To Link Criminal Justice Reentrants To Hiv Care In The Community, Ann Kurth, Irene Kuo, James Peterson, Nkiru Azikiwe, Lauri Bazerman, Alice Cates, Curt G. Beckwith

Epidemiology Faculty Publications

The United States has the world’s highest prison population, and an estimated one in seven HIV-positive persons in the USA passes through a correctional facility annually. Given this, it is critical to develop innovative and effective approaches to support HIV treatment and retention in care among HIV-positive individuals involved in the criminal justice (CJ) system. Information and communication technologies (ICTs), including mobile health (mHealth) interventions, may offer one component of a successful strategy for linkage/retention in care. We describe CARE+ Corrections, a randomized controlled trial (RCT) study now underway in Washington, that will evaluate the combined effect of computerized motivational …

Complex Dynamics In Predator-Prey Models With Nonmonotonic Functional Response And Seasonal Harvesting, Jicai Huang, Jing Chen, Yijun Gong, Weipeng Zhang

Mathematics Faculty Articles

In this paper we study the complex dynamics of predator-prey systems with nonmonotonic functional response and harvesting. When the harvesting is constant-yield for prey, it is shown that various kinds of bifurcations, such as saddle-node bifurcation, degenerate Hopf bifurcation, and Bogdanov-Takens bifurcation, occur in the model as parameters vary. The existence of two limit cycles and a homoclinic loop is established by numerical simulations. When the harvesting is seasonal for both species, sufficient conditions for the existence of an asymptotically stable periodic solution and bifurcation of a stable periodic orbit into a stable invariant torus of the model are given. …

Sources Of The Persistent Gender Wage Gap Along The Unconditional Earnings Distribution: Findings From Kenya, Richard U. Agesa, Jacqueline Agesa, Andrew Dabalen

Economics Faculty Research

Past studies on gender wage inequality in Africa typically attribute the gender pay gap either to gender differences in characteristics or in the return to characteristics. The authors suggest, however, that this understanding of the two sources may be far too general and possibly overlook the underlying covariates that drive the gender wage gap. Moreover, past studies focus on the gender wage gap exclusively at the conditional mean. The authors go further to evaluate the partial contribution of each wage-determining covariate to the magnitude of the gender pay gap along the unconditional earnings distribution. The authors' data are from Kenya, …

Determinants Of Negative Pathways To Care And Their Impact On Service Disengagement In First-Episode Psychosis., Kelly K. Anderson, Rebecca Fuhrer, Norbert Schmitz, Ashok K Malla

Epidemiology and Biostatistics Publications

PURPOSE: Although there have been numerous studies on pathways to care in first-episode psychosis (FEP), few have examined the determinants of the pathway to care and its impact on subsequent engagement with mental health services.

METHODS: Using a sample of 324 FEP patients from a catchment area-based early intervention (EI) program in Montréal, we estimated the association of several socio-demographic, clinical, and service-level factors with negative pathways to care and treatment delay. We also assessed the impact of the pathway to care on time to disengagement from EI services.

RESULTS: Few socio-demographic or clinical factors were predictive of negative pathways …

Comparison Of Molecular Breeding Values Based On Within- And Across-Breed Training In Beef Cattle, Stephen D. Kachman, Matthew L. Spangler, Gary L. Bennett, Kathryn J. Hanford, Larry A. Kuehn, Warren M. Snelling, Mark Thallman, Mahdi Saatchi, Dorian J. Garrick, Robert D. Schnabel, Jeremy F. Taylor, E John Pollak

Department of Statistics: Faculty Publications

Background: Although the efficacy of genomic predictors based on within-breed training looks promising, it is necessary to develop and evaluate across-breed predictors for the technology to be fully applied in the beef industry. The efficacies of genomic predictors trained in one breed and utilized to predict genetic merit in differing breeds based on simulation studies have been reported, as have the efficacies of predictors trained using data from multiple breeds to predict the genetic merit of purebreds. However, comparable studies using beef cattle field data have not been reported.

Methods: Molecular breeding values for weaning and yearling weight …

Obesity And Preference-Weighted Quality Of Life Of Ethnically Diverse Middle School Children: The Healthy Study, Roberto P. Trevino, Trang H. Pham, Sharon Edelstein

GW Biostatistics Center

No abstract provided.

Transmission Potential Of Influenza A/H7n9, February To May 2013, China, Gerardo Chowell, Lone Simonsen, Sherry Towers, Mark A. Miller, Cecile G. Viboud

Epidemiology Faculty Publications

Background

On 31 March 2013, the first human infections with the novel influenza A/H7N9 virus were reported in Eastern China. The outbreak expanded rapidly in geographic scope and size, with a total of 132 laboratory-confirmed cases reported by 3 June 2013, in 10 Chinese provinces and Taiwan. The incidence of A/H7N9 cases has stalled in recent weeks, presumably as a consequence of live bird market closures in the most heavily affected areas. Here we compare the transmission potential of influenza A/H7N9 with that of other emerging pathogens and evaluate the impact of intervention measures in an effort to guide pandemic …

A Case–Control Study Of Incident Rheumatological Conditions Following Acute Gastroenteritis During Military Deployment, Kathryn Deyoung, Mark A. Riddle, Larissa S. May, Chad K. Porter

Epidemiology Faculty Publications

Objectives The aim of this study was to assess the risk of incident rheumatological diagnoses (RD) associated with self-reported diarrhoea and vomiting during a first-time deployment to Iraq or Afghanistan. Such an association would provide evidence that RD in this population may include individuals with reactive arthritis (ReA) from deployment-related infectious gastroenteritis.

Design This case–control epidemiological study used univariate and multivariate logistic regression to compare the odds of self-reported diarrhoea/vomiting among deployed US military personnel with incident RD to the odds of diarrhoea/vomiting among a control population.

Setting We analysed health records of personnel deployed to Iraq or Afghanistan, including …

Statistical Analysis Of Microarray Data In Sleep Deprivation, Stephanie Marie Berhorst

Masters Theses

"Microarray technology is a useful tool for studying the expression levels of thousands of genes or exons within a single experiment. Data from microarray experiments present many challenges for researchers since costly resources often limit the experimenter to small sample sizes and large amounts of data are generated. The researcher must carefully consider the appropriate statistical analysis to use that aligns with the experimental design employed. In this work, statistical issues are investigated and addressed for a microarray experiment that examines how expression levels change over time as individuals are sleep deprived. Over the course of 48 hours of sleep …

Estimation And Q-Matrix Validation For Diagnostic Classification Models, Yuling Feng

Theses and Dissertations

Diagnostic classification models (DCMs) are structured latent class models widely discussed in the field of psychometrics. They model subjects' underlying attribute patterns and classify subjects into unobservable groups based on their mastery of attributes required to answer the items correctly. The effective implementation of DCMs depends on correct specification of a Q-matrix which is a binary matrix linking attribute patterns to items. Current literature on assessing the appropriateness of Q-matrix specifications has focused on validation methods for the deterministic-input, noisy-and-gate (DINA) model. The goal of the study is to develop general Q-matrix validation methods that can be applied to a …

Heaped Data In Count Models, Tammy Harris

Theses and Dissertations

Heaped data result when subjects who recall the frequency of events prefer for reporting from a limited set of rounded responses or preferred digits over reporting exact counts. These rounded responses and digit preferences (also referred to as data coarsening) could be characterized by reported frequencies (or counts) favoring multiples of 20, reporting counts ending with 0 or 5, or a preference for reporting an even number over an odd number or vice versa. This mixture of values is a type of measurement error (pattern of misreporting) that can lead to biased estimation and imprecision in discrete quantitative data. Sometimes …

The Psychological Impacts Of False Positive Ovarian Cancer Screening: Assessment Via Mixed And Trajectory Modeling, Amanda T. Wiggins

Theses and Dissertations--Epidemiology and Biostatistics

Ovarian cancer (OC) is the fifth most common cancer among women and has the highest mortality of any cancer of the female reproductive system. The majority (61%) of OC cases are diagnosed at a distant stage. Because diagnoses occur most commonly at a late-stage and prognosis for advanced disease is poor, research focusing on the development of effective OC screening methods to facilitate early detection in high-risk, asymptomatic women is fundamental in reducing OC-specific mortality. Presently, there is no screening modality proven efficacious in reducing OC-mortality. However, transvaginal ultrasonography (TVS) has shown value in early detection of OC. TVS presents …

New Microarray Image Segmentation Using Segmentation Based Contours Method, Yuan Cheng

Doctoral Dissertations

The goal of the research developed in this dissertation is to develop a more accurate segmentation method for Affymetrix microarray images. The Affymetrix microarray biotechnologies have become increasingly important in the biomedical research field. Affymetrix microarray images are widely used in disease diagnostics and disease control. They are capable of monitoring the expression levels of thousands of genes simultaneously. Hence, scientists can get a deep understanding on genomic regulation, interaction and expression by using such tools.

We also introduce a novel Affymetrix microarray image simulation model and how the Affymetrix microarray image is simulated by using this model. This simulation …

The Impact Of Local Patent Rules On Rate And Timing Of Case Resolution Relative To Claim Construction: An Empirical Study Of The Past Decade, Pauline M. Pelletier

Journal of Business & Technology Law

No abstract provided.

Numerical Solutions To The Gross-Pitaevskii Equation For Bose-Einstein Condensates, Luigi Galati

College of Graduate Studies: Theses & Dissertations

In this thesis we compare various potential operators for the two-dimensional (2D) Gross-Pitaevskii equation (GPE) for Bose-Einstein condensates. Both the 2D and the 1D models are scaled to get a three parameter model. Smoothness of initial conditions is considered and choice of method (Split-Step Fourier method with Strang Splitting) is justied. Numerical simulations provide graphical evidence of properties of both focusing and nonfocusing cases.

Nfl Betting Market: Using Adjusted Statistics To Test Market Efficiency And Build A Betting Model, James P. Donnelly

CMC Senior Theses

The use of statistical analysis has been prevalent in the sports gambling industry for years. More recently, we have seen the emergence of "adjusted statistics", a more sophisticated way to examine each play and each result (further explanation below). And while adjusted statistics have become commonplace for professional and recreational bettors alike, little research has been done to justify their use. In this paper the effectiveness of this data is tested on the most heavily wagered sport in the world – the National Football League (NFL). The results are studied with two central questions in mind: Does the market account …

Why Police Learn From Third-Party Data, Randall K. Johnson

Faculty Works

This essay argues that third-party data collection, particularly of administrative complaints and departmental audit information, holds greater promise than lawsuit data collection. It does so by asserting that third-party data collection is more useful for three reasons. First, third-party data collection prevents manipulation by individual police officers and law enforcement agencies. Second, it assures that police behavioral trends are actually identified. Lastly, third-party data collection helps to deter published § 1983 cases. The essay, however, only models and tests the final claim.

Connecting Big Data With Big Decisions: Ideas For Synthesizing Analytics And Decision Analysis, Jeffrey Keisler

Management Science and Information Systems Faculty Publication Series

This paper describes an approach to connect decision analysis models with outputs of analytic methods applied to various types of big data. Decision analysis models focus on issues of concern to a decision maker and incorporate use of a range of methods and axioms to develop insights about what the decision maker should do. In particular, decision analysis models typically use subjective judgments from the decision maker to describe beliefs about the likelihood of events and the desirability of outcomes. In order for human judgments to be improved by the availability of large amounts of data and processing power, it …

The Spectral Analysis Of Longitudinal Data With Applications To Atmospheric Variables, Amy Potrzeba Macrina

Legacy Theses & Dissertations (2009 - 2024)

Longitudinal data analysis is an observational study that analyzes data that has been collected over long periods of time where time can be arbitrary. It is a popular method that has been widely used over the time domain approach. An equivalent method to the time domain approach is the frequency domain approach. This method allows researchers to observe the spectral properties of the data. In the world everything exists in continuous time t, but we made observations of a given variable at some specific times. Currently spectral analysis is well developed for time series when observations are made at equal …

Raman Spectroscopy For The Identification Of Body Fluid Traces : Mixtures And Contaminations, Race And Gender Differentation, Aliaksandra Sikirzhytskaya

Legacy Theses & Dissertations (2009 - 2024)

Body fluid traces are an important type of forensic evidence, which play a significant role in the reconstruction of a violent crime and often help to identify a victim or suspect based on DNA analysis. Despite a great need, there is no a single method, which could identify multiple body fluids. In addition, majority of current methods are destructive for the evidence. For about eight years, our laboratory has been working on the development of a new nondestructive method for identification of body fluid traces based on Raman microspectroscopy combined with advanced statistics. High differentiation power of the method has …

Non-Likelihood Based Model Evaluation And Comparison With Application To Genetic And Clinical Hiv-1 Outcomes, Ashley Elise Giambrone

Legacy Theses & Dissertations (2009 - 2024)

Although treatment for human immunodeficiency virus type-1 (HIV-1) has undergone drastic change and morbidity and mortality has decreased over time, the development of drug-resistant HIV-1 is of concern for the long-term antiretroviral treatment of infected individuals. Drug-resistant virus is known to manifest with potentially complex mutational patterns in the HIV-1 genotype sequence and is associated with decreased response to therapy. Resistance occurs either as a result of development of mutations in the viral genome under selective drug pressure or as a result of naturally occurring polymorphisms. The most effective treatment methods are still debated at this time; however, current treatment …

Augmenting Program Data With Secondary Data Sources To Improve The Quality Of Existing Statistics : Four Examples From The New York State Department Of Health Hospital-Acquired Infection Reporting Program, Valerie Benson Haley

Legacy Theses & Dissertations (2009 - 2024)

The objective of this dissertation is to illustrate the novel application of methods that can be used to improve the accuracy of hospital-acquired infection (HAI) rates. Public reporting of HAI rates is relatively new. New York was one of the first states to mandate reporting in acute care hospitals (2007), followed by national pay-for-reporting in 2011, and national value-based purchasing in 2013. Given the financial ramifications of public reporting, it is critical that the data are validated and adjusted for differences in underlying risk among patient populations so that hospital performance can be fairly compared. However, limited information on the …

Flexible Variable And Model Selection With Ordinal Categorical Responses And Multiple Covariates, Wan-Hsiang Hsu

Legacy Theses & Dissertations (2009 - 2024)

Ordered categorical responses are common in many applied studies; moreover, with the rapid growth of computational power and technologies, ultrahigh dimensional data becomes very widespread. For example, there could be ten thousands of dimensions in a gene expression data and of interest is to classify the disease stage with specific genes and predict a clinical prognosis by using these specific genes. However, there are some unique challenges, including (1) the curse of dimensionality and (2) the modeling strategy for allowing dynamic covariate effects. Thus, variable selection for mining ultrahigh dimensional data and flexible modeling strategy for allowing dynamic covariate effects …

Data Mining And Pattern Discovery Using Exploratory And Visualization Methods For Large Multidimensional Datasets, Hsin-Fang Li

Theses and Dissertations--Epidemiology and Biostatistics

Oral health problems have been a major public health concern profoundly affecting people’s general health and quality of life. Given that oral health data is composed of several measurable dimensions including clinical measurements, socio-behavioral factors, genetic predispositions, self-reported assessments, and quality of life measures, strategies for analyzing multidimensional data are neither computationally straightforward nor efficient. Researchers face major challenges to identify tools that circumvent the processes of manually probing the data.

The purpose of this dissertation is to provide applications of the proposed methodology on oral health-related data that go beyond identifying risk factors from a single dimension, and to …