Statistics and Probability Commons^™

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 …

Quantifying The Risks Of Soil Lead In Urban Community Gardens: Sampling To Account For Spatial Variability, Lauren Elizabeth-Korgol Bugdalski

Wayne State University Theses

Urban gardening has recently gained popularity as a way to provide fresh produce and income to urban residents; however, finding suitable sites for urban gardens is challenging because of historical soil lead contamination particularly in post- industrial cities like Detroit, Michigan. Soil lead measurements from three Detroit gardens were modeled using geostatistical techniques to assess risk and alternate sampling strategies. General sampling recommendations for urban gardens were developed based on results of Monte Carlo simulations and associated risk assessment. Variograms and kriged concentration maps indicate spatial variability at scales as small as one meter, with site specific variability in spatial …

Undergraduate And Graduate Teaching Assistants' Perceptions Of Their Responsibilities - Factors That Help Or Hinder, Alena Moon, Hyunyi Jung, Farshid Marbouti, Kelsey Joy Rodgers, Heidi A. Diefes-Dux

Mathematics, Statistics and Computer Science Faculty Research and Publications

Effective teaching assistants (TAs) are crucial for effective student learning. This is especially true in science, technology, engineering, and mathematics (STEM) programs, where TAs are enabling large programs to transition to more student-centered learning environments. To ensure that TAs are able to support these types of learning environments, their perspectives of training, their abilities, and other work related aspects must be understood. In this paper a survey that was created based on interviews conducted with eight TAs is discussed. The survey has four primary categories of content that are critical for understanding TAs' perspectives: (1) background, (2) motivation, (3) training, …

Genome-Wide Associations Of Signaling Pathways In Glioblastoma Multiforme, Serdar Bozdag, Stefan Wuchty, Alexei Vazquez, Peter O. Bauer

Mathematics, Statistics and Computer Science Faculty Research and Publications

Background: eQTL analysis is a powerful method that allows the identification of causal genomic alterations, providing an explanation of expression changes of single genes. However, genes mediate their biological roles in groups rather than in isolation, prompting us to extend the concept of eQTLs to whole gene pathways. Methods: We combined matched genomic alteration and gene expression data of glioblastoma patients and determined associations between the expression of signaling pathways and genomic copy number alterations with a non-linear machine learning approach. Results: Expectedly, over-expressed pathways were largely associated to tag-loci on chromosomes with signature alterations. Surprisingly, tag-loci that were associated …

A Proposal For A Problem-Driven Mathematics Curriculum Framework, Judith S. Zawojewski, Marta Magiera, Richard Lesh

Mathematics, Statistics and Computer Science Faculty Research and Publications

A framework for a problem-driven mathematics curriculum is proposed, grounded in the assumption that students learn mathematics while engaged in complex problem-solving activity. The framework is envisioned as a dynamic technologicallydriven multi-dimensional representation that can highlight the nature of the curriculum (e.g., revealing the relationship among modeling, conceptual, and procedural knowledge), can be used for programmatic, classroom and individual assessment, and can be easily revised to reflect ongoing changes in disciplinary knowledge development and important applications of mathematics. The discussion prompts ideas and questions for future development of the envisioned software needed to enact such a framework.

Rssi Based Indoor Localization For Smartphone Using Fixed And Mobile Wireless Node, Md O. Gani, Casey O'Brien, Sheikh Iqbal Ahamed, Roger O. Smith

Mathematics, Statistics and Computer Science Faculty Research and Publications

Nowadays with the dispersion of wireless networks, smartphones and diverse related services, different localization techniques have been developed. Global Positioning System (GPS) has a high rate of accuracy for outdoor localization but the signal is not available inside of buildings. Also other existing methods for indoor localization have low accuracy. In addition, they use fixed infrastructure support. In this paper, we present a novel system for indoor localization, which also works well outside. We have developed a mathematical model for estimating location (distance and direction) of a mobile device using wireless technology. Our experimental results on Smartphones (Android and iOS) …

Recent Advances In Energy-Efficient Sensor Networks, Sheikh Iqbal Ahamed, Wei Wang, Jiman Hong

Mathematics, Statistics and Computer Science Faculty Research and Publications

No abstract provided.

Concept Of Sub-Independence, Gholamhossein Hamedani

Mathematics, Statistics and Computer Science Faculty Research and Publications

Limit theorems as well as other well-known results in probability and statistics are often based on the distributions of the sums of independent random variables. The concept of sub-independence, which is weaker than that of independence, is shown to be sufficient to yield the conclusions of these theorems and results. It also provides a measure of dissociation between two random variables which is stronger than uncorrelatedness.

Characterizations Of Exponentiated Distributions, Gholamhossein Hamedani

Mathematics, Statistics and Computer Science Faculty Research and Publications

Various characterizations of the class of exponentiated distributions are presented. These characterizations are based on a simple relationship between two truncated moments and based on functions of the nth order statistic. The results are applied to certain well-known members of this class.

Robust Estimation Of The Correlation Matrix Of Longitudinal Data, Mehdi Maadooliat, Mohsen Pourahmadi, Jianhua Z. Huang

Mathematics, Statistics and Computer Science Faculty Research and Publications

We propose a double-robust procedure for modeling the correlation matrix of a longitudinal dataset. It is based on an alternative Cholesky decomposition of the form Σ=DLL ^⊤ D where D is a diagonal matrix proportional to the square roots of the diagonal entries of Σ and L is a unit lower-triangular matrix determining solely the correlation matrix. The first robustness is with respect to model misspecification for the innovation variances in D, and the second is robustness to outliers in the data. The latter is handled using heavy-tailed multivariate t-distributions with unknown degrees of freedom. We …

The Zografos-Balakrishnan Log-Logistic Distribution: Properties And Applications, Gholamhossein Hamedani

Mathematics, Statistics and Computer Science Faculty Research and Publications

No abstract provided.

Cross‐Campus Collaboration: A Scientometric And Network Case Study Of Publication Activity Across Two Campuses Of A Single Institution, Jeremy Birnholtz, Shion Guha, Y. Connie Yuan, Geri Gay, Caren Heller

Mathematics, Statistics and Computer Science Faculty Research and Publications

Team science and collaboration have become crucial to addressing key research questions confronting society. Institutions that are spread across multiple geographic locations face additional challenges. To better understand the nature of cross‐campus collaboration within a single institution and the effects of institutional efforts to spark collaboration, we conducted a case study of collaboration at Cornell University using scientometric and network analyses. Results suggest that cross‐campus collaboration is increasingly common, but is accounted for primarily by a relatively small number of departments and individual researchers. Specific researchers involved in many collaborative projects are identified, and their unique characteristics are described. Institutional …

Student Fact Book, Fall 2013, Thirty-Seventh Annual Edition, Wright State University, Office Of Student Information Systems, Wright State University

Wright State University Student Fact Books

The student fact book has general demographic information on all students enrolled at Wright State University for Fall Semester, 2013.