Applied Statistics Commons^™

Group Antenatal Care Positively Transforms The Care Experience: Results Of An Effectiveness Trial In Malawi, Crystal L. Patil, Kathleen F. Noor, Esnath Kapito, Li C. Liu, Xiaohan Mei, Elizabeth Chodzaza, Genesis Chorwe-Sungani, Ursula Kafuulafula, Elizabeth T. Abrams, Allissa Desloge, Ashley Gresh, Rohan D. Jeremiah, Dhruvi R. Patel, Anne Batchelder, Heidy Wang, Jocelyn Faydenko, Sharon S. Rising, Ellen Chirwa

Department of Medicine Faculty Publications

Background

We developed and tested a Centering-based group antenatal (ANC) model in Malawi, integrating health promotion for HIV prevention and mental health. We present effectiveness data and examine congruence with only the Group ANC theory of change model, which identifies key processes as supportive relationships, empowered partners in learning and care, and meaningful services, leading to better ANC experiences and outcomes.

Methods

We conducted a hybrid effectiveness-implementation trial at seven clinics in Blantyre District, Malawi, comparing outcomes for 1887 pregnant women randomly assigned to Group ANC or Individual ANC. Group effects on outcomes were summarized and evaluated using t-tests, Mann-Whitney, …

Machine Learning Methods For Intrusion Detection And Response In Network Security, Ayomide Oyemaja

College of Graduate Studies: Theses & Dissertations

Intrusion Detection Systems (IDS) play a crucial role in computer network security by identifying malicious activities and potential cyberattacks. This thesis combines machine learning and cybersecurity by applying Reinforcement Learning (RL) in intrusion detection and response using the NSL-KDD dataset.

We designed and implemented a Q-learning framework where an agent learns to classify network traffic over time by interacting with the environment and receiving rewards based on detection accuracy. We also look at the importance of feature selection and classification techniques and how effective they are in improving model performance, reducing the complexity of computation, and producing more desirable results. …

Majority Decision Using Top-Performing Neural Networks Models For Improved Credit Risk Prediction, Vincent Dey

College of Graduate Studies: Theses & Dissertations

Credit risk prediction remains both a challenging and high-interest problem due to the inherently unbalanced nature of financial datasets and the continuous drive for higher pre- dictive precision. In this work, I build upon previous advancements in credit risk modeling and introduce an ensemble-based Artificial Neural Network (ANN) architecture designed to enhance classification performance. By leveraging a selective ensemble of decision net- works, this approach not only improves prediction accuracy but also mitigates the chal- lenges posed by imbalanced data distributions. While the primary focus is on credit risk prediction, my analysis demonstrates that the proposed model can be effectively …

Theoretical Foundations And Applied Performance Of Periodicity-Aware Imputation: Variable Bandpass Block Bootstrap Methods For Incomplete Time Series, Asmaa Ahmad

Electronic Theses & Dissertations (2024 - present)

Time series data are prevalent across a wide range of disciplines, including health surveillance, public policy, and environmental monitoring. In the presence of underlying cyclical patterns, the integrity of time series analysis depends critically on the ability to detect, model, and impute structured missing data without compromising the temporal structure. This dissertation introduces and validates a novel imputation framework that integrates the Variable Bandpass Periodic Block Bootstrap (VBPBB) into multiple imputation procedures, improving the accuracy, robustness, and interpretability of time series models under high rates of missingness and noise. The overarching goal of this dissertation was to develop and evaluate …

Mathematical Contributions To The Study Of Chemotaxis And Cell Signaling, Hajr Zam

Graduate Theses, Dissertations, and Problem Reports (ETD)

This dissertation presents results from two mathematical projects concerned with the biology of cells. Chapter 1 provides biological background and places the two mathematical problems in the context of cell signaling. The larger project, with Prof. H. Hattori on a chemotaxis model is presented in Chapters 3 and 4. Work with Prof. \'{A}. Hal\'{a}sz on a chemical reaction network system with linear multimers and two types of labels is presented in Chapter 2. The chemotaxis system describes the one-dimensional dynamics of a species of cells with two chemical species, a chemo-attractant and chemo-repellent. The goal is to analyze the behavior …

Analyzing Factors Influencing Employee Turnover In Tech Companies: A Predictive Modeling Approach, Shinjon Ghosh

Theses and Dissertations

Employee turnover poses substantial challenges for technology firms, and understanding its key drivers through predictive modeling is essential for developing effective retention strategies. This study investigates factors influencing employee turnover in technology companies by implementing a predictive modeling approach on the IBM HR Analytics Employee Attrition dataset. The research aims were identifying key factors contributing to employee attrition, developing predictive models to forecast turnover risk, and analyzing interactions among significant predictors. By examining a range of features, the results highlight significant variables (Over Time, Monthly Income, Marital Status, etc.) of attrition and offer actionable insights for developing targeted employee retention …

Predictive Inference For Ion Concentration With Machine Learning And Bayesian Methods, Alexandra B. Ulbing

Theses and Dissertations

Ultraviolet--visible (UV--Vis) spectroscopy produces high-dimensional signals that are strongly collinear, shift with concentration, and exhibit heteroskedastic, non-Gaussian noise. These features make supervised regression from spectra to ionic concentrations statistically challenging and limit the reliability of methods that assume linear structure or homoscedastic errors.

This dissertation develops two complementary frameworks for prediction and uncertainty quantification in UV--Vis spectroscopic regression: (1) frequentist stacked ensembles combined with distribution-free conformal prediction, and (2) Bayesian hierarchical modeling and Bayesian stacking. Together, they provide a unified view of model-based and distribution-free uncertainty across nickel and nickel--cobalt datasets.

The frequentist component builds ensembles of Functional Data Analysis …

Impact Of Urban Development On Uv Exposure: A Clustering And Machine Learning Assessment, Taufik Roni Sahroni Mr., Verdi Yasin, Lulut Alfaris, Reza Ariefka, Ruben Cornelius Siagian, Mohammad Alfin Karim, Nana Rahdiana, Ade Suhara

Journal of Environmental Science and Sustainable Development

The relocation of Indonesia's capital city is anticipated to promote inclusive economic growth while embracing cultural diversity. However, this transition may affect ultraviolet (UV) radiation exposure patterns. The study investigated variations in UV exposure in the IKN region, focusing on urban development factors such as land use and population density that affect public health, sun protection, and skin cancer prevention. The research hypothesized that UV radiation is significantly correlated with these factors. UV Index data from 2010-2023, a hierarchical clustering method, identifies complex data patterns without determining the number of clusters. XGBoost, a machine learning model, was used for handling …

Mean From Median Estimation In Longitudinal Meta-Analysis Of Quality Of Life In Radiation Oncology Patients, Harlan R. Sayles

Theses & Dissertations

Meta-analysis of longitudinal data, where some of the studies report results with means and standard errors while others use medians and ranges, is a complex analytical problem with several challenges which must be overcome. Existing methods for estimating means from medians and either ranges, interquartile ranges, or both have not previously been evaluated in a longitudinal setting. In this work, a simulation study was used to estimate mean bias, between studies variance bias, and coverage of confidence intervals in a longitudinal setting. A second simulation study estimated the variance of meta-analysis results from a given set of studies that may …

Calculation And Statistical Analysis Of Wins Above Replacement, Joshua Taylor

Departmental Honors & Graduate Capstone Projects

The Wins Above Replacement (WAR) statistic in Major League Baseball is a prominent metric used to estimate player value by quantifying all aspects of play in terms of wins added to a baseball team. We will use R to calculate WAR for all players from 1871 to 2012 and use data from those years to construct multivariate predictive models to attempt to estimate WAR for players from 2013 to 2024. We find strong correlations between predicted and actual WAR values for most models, with the exception of the polynomial predictive model for non-qualified pitchers.

(R2095) On Discrete Hypoexponential Distribution And Integer-Valued Autoregressive Process, K. Krishnakumari, Dais George

Applications and Applied Mathematics: An International Journal (AAM)

Discrete distributions provide a substantive contribution in modeling real world frame work. Though a lot of discrete distributions are available in literature, they are inappropriate to model many practical situations. The conventional discrete distributions like geometric, Negative Binomial and Poisson have limited applications in modeling count and lifetime data. This paper introduces a new two parameter discrete distribution namely, discrete hypoexponential distribution by discretizing the well known hypoexponential distribution. Various distributional and structural properties of the proposed distribution are studied. We introduce a first order auto regressive process with the newly proposed distribution as marginal and study the properties. Depending …

(R2116) The Novel Garima Distribution Properties And Nuclear Data Application, Murat Aygün, Ayşe Metin Karakaş

Applications and Applied Mathematics: An International Journal (AAM)

This research introduces a new two-parameter Marshall-Olkin Garima distribution model. The novel model has many sub-models that are useful in modeling real-life data, such as the extended Garima distribution, exponentiated Garima distribution, exponential distribution, Lindley distribution, Kumaraswamy Garima distribution, and normal distribution. The proposed model demonstrates a high level of suitability in modeling both reliability and survival data. It is flexible in accommodating various failures. The quantile function, density shapes, hazard rate functions, and order statistics are a few of the statistical features that have been explored. Maximum likelihood estimation methods were employed to estimate the parameters. Using five data …

Comparison Of Statistical And Machine Learning Genomic Prediction Methods In Plant Breeding: Case Studies In Maize And Soybean, Igor Kuivjogi Fernandes

Graduate Theses and Dissertations

Plant breeding is essential to increase genetic gain and food production worldwide. This study was conducted to evaluate new ways to use machine learning (ML) to tackle plant breeding challenges, where two ideas were tested — the first chapter focuses on how to combine genetic and environmental data using ML to improve the prediction of maize grain yield in multi-environment trials, while the second chapter centers on how to couple feature selection of molecular markers with ML to enhance prediction of yield in soybean, and, in both cases, ML approaches were compared to well-established statistical methods greatly adopted by the …

Policies And Price Controls On The Research And Development Of Orphan Drugs In The United States And The European Union, Bena Pearl Filipczak Smith

Master's Theses

There is substantive literature surrounding the impact of price controls on the research and development (R&D) of new pharmaceutical products. The European Union (EU) and United States (US) are often studied in contrast to examine the influence of price controls as the US has fewer pharmaceutical price controls than the EU.

We find moderate evidence that the US spent more on annual domestic pharmaceutical R&D than the EU between 2004 and 2021, on average, before and after adjusting for GDP growth per capita and year. We find strong evidence that the US increased annual domestic R&D spending at a faster …

Quantile Regression And Change Point Analysis Of Remote Patient Monitoring Data From Cardiomems Hf System, Shanshan Jia

All Dissertations

Quantile regression provides a sophisticated analytical approach for clinical data, offering deeper insights into patient variability and risk assessment than conventional regression techniques. This study delves into the mathematical underpinnings of quantile regression, highlighting its advantages over ordinary least squares (OLS) regression, particularly in the context of predicting diastolic pulmonary artery pressure (PAP). We explore how this method can be applied to guide clinical interventions more effectively. By leveraging quantile regression’s ability to model different parts of the outcome distribution, we demonstrate its potential to enhance patient care through improved identification of high-risk individuals and the development of more personalized …

Striking A Balance: Market Shock & Responses In Automotive Components Manufacturing, Emma Lane Mcgahey

All Theses

This thesis examines the effects of extreme market shocks on supply chain dynamics within the automotive industry. Through an analysis of demand data from an automotive manufacturer to its component suppliers (January 2018 to May 2024), the study investigates the relationship between market shocks and supply chain responses, providing insights into how auto components inventory management handles downstream responses to market shocks. With supporting public data—from FRED, BLS, and the U.S. Census Bureau resources—we explore two primary relationships: the impact of market shocks on the Average Standard Deviation of Demand (SDO) and the effect of demand variability on expedited pricing …

A Strategic Insight Into The Market For Carbon Management Capacity, Mahelet G. Fikru, Ting Shen, Jennifer Brodmann, Hongyan Ma

Economics Faculty Research & Creative Works

This study presents the market for carbon management capacity via carbon capture, utilization, and storage technologies, identifying demand and supply forces, as well as clarifying the potential impact of market and non-market-based shocks on technology developers versus adopters. The paper addresses a prevailing gap in market analysis, introducing a microeconomic framework and unique dataset to identify key players, market forces, and policy incentives shaping the carbon capture, utilization, and storage landscape. The analysis equips industry stakeholders, policymakers, and investors with valuable insights regarding (1) leaders in the design, development, and manufacture of carbon capture, utilization, and storage technologies (supply), (2) …

Financialization And Price Volatility: An Empirical Analysis On Speculation In The Oil Market, Audry F. Oliveira Carnivale

Electronic Theses and Dissertations

Financialization has facilitated the trade of futures contracts because of deregulatory policies increasing speculation. Speculation has created a more fragile market inducing riskier investments and aggravating price volatility. Three post-Keynesian theories, the financial instability hypothesis, money manager capitalism and markup, explain how policy altering the banking structure has developed financialization from lax regulation. Previous research has emphasized supply and demand as the main determinants of oil price changes, but it’s important to consider how structural changes from policy stimulating a more financialized economy has impacted volatility. With Brent Crude oil price data and West Texas Intermediate (WTI) open interest and …

Cure Rate Analysis Of National Health Insurance Scheme Claims Payment Survival Times In Ghana: The Case Of Pru District, Ahmed Tamimu, Suleman Nasiru, Dioggban Jakperik

Al-Bahir

In this study, cure models have been applied to model National Health Insurance Scheme (NHIS) claims payment data with cured proportion using Pru District in the Bono East Region of Ghana as a case study. The covariates effects were also modelled to investigate the effects of the covariates on the cured proportion. The estimates of the parameters of the models were obtained by directly maximizing the observed likelihood functions. Most estimates of the parameters of the cure models are significant at 5% significance level. The study revealed that the cured claims payments rate increases over time with an estimated cured …

Supplementary Files For: Quantifying The Impact Of Rain-On-Snow Induced Flooding In The Western United States, Emma Watts, Brennan Bean

Browse all Datasets

Serious flooding can happen when rain falls on snow, which we call a rain-on-snow (ROS) event. Increasing our understanding of the behavior of floods resulting from ROS events can help us design better systems to manage flood water and prevent it from causing damage. This thesis explores how ROS events affect streamflow in the Western United States by examining the weather conditions that precede a streamflow surge. We classify stream surges as ROS or non-ROS induced based on these weather conditions, which helps us separate floods caused by ROS events from those caused by other factors. By comparing these different …