Statistical Methodology Commons^™

A Geometric Framework For Modeling Real-World Experience, Knowledge, Learning, And Memory, Paxton C. Fitzpatrick

Dartmouth College Ph.D Dissertations

A growing body of recent work highlights the importance of studying learning and memory in “naturalistic” contexts that reflect how these processes function in our day-to-day lives. Yet the incredible complexity of the world around us, and of the idiosyncratic ways we learn from and remember it, make characterizing the dynamics of learning and memory for real-world experiences challenging. I present a computational framework that leverages the embedding representations of language models to cast experiences and memories as high-dimensional trajectories unfolding through a common representation space, enabling them to be directly related and compared via their latent semantic geometries. I …

Behind The Score: The Influence Of Neighborhood Policy Contexts On New York Regents Exam Performance Across New York City Public High Schools, Philip J. Kovacevic

Dissertations, Theses, and Capstone Projects

This study used a two-level hierarchical linear model (HLM) to test whether New York City neighborhood zoning and land-use characteristics independently predict school-level performance on the New York State Regents Examinations, after controlling for school demographic and organizational characteristics, and Regents mean scores, both drawn from the NYC DOE's Regents Exam Results file (NYC Public Schools InfoHub, 2024). Using NYC Primary Land Use Tax Lot Output (PLUTO) 2025 data, a Neighborhood Educational Quality (NEQ) index was built for each of the city’s 42 community districts by combining the commercial floor area ratio (CommFAR) and the proportion of lots zoned residential …

Modeling Mean And Variability Of Anxiety In Ecological Momentary Assessment Data Using Mixed-Effects Location–Scale Models, Trenzy Odero

Electronic Theses and Dissertations

Ecological Momentary Assessment is a method of collecting repeated measures of people in real time within natural environments. This results in hierarchical data that has a significant amount of variation at the person level. The traditional linear mixedeffects models assume that the residual variance is constant, which might not be true when the residual variance varies among individuals as well as in time. This thesis uses mixed-effects location-scale (MELS) models to model the mean and variance of an EMA outcome together. By introducing the possibility of variability in residual variance within and across individuals and with covariates, the MELS framework …

Bias, Structure, And Inference In Applied Network Analysis, Anna Vasenina

Dartmouth College Ph.D Dissertations

This dissertation develops mathematical and statistical methods for extracting reliable information from network data across biological applications, with an emphasis on understanding what observed network structure can and cannot resolve. The first study leverages protein–protein interaction network topology in the c-di-GMP signaling system of Pseudomonas fluorescens, showing that node centrality measures accurately classify protein domain types and that physical interaction structure contributes statistically significant predictive power for biofilm formation phenotypes across nearly 200 environments, while gene expression does not. The second study examines sampling bias in lemur-plant trophic interaction networks in Madagascar, demonstrating that differential detection of diurnal versus …

Developing A Humpback Whale Vocalization Detector Using Machine Learning Models, Lucas Kantorowski

Master's Theses

Humpback whale songs are notoriously complex. Identification of humpback whale song units requires bioacousticians to tediously listen, analyze, and annotate collected sound data. Even sparse data requires listening to the entirety of the collected acoustic data. In this study, three hours of audio containing over one-thousand humpback whale song units was collected in Monterey Bay, California.

Prior studies have seen success using convolutional neural networks by performing image classification on hundreds of hours worth of spectrograms. Our study uses traditional machine learning models, as they are less computationally demanding, and require less data.

We use time splitting and Mel-frequency cepstrum …

A Predictive Coding Account Of Spatial Working Memory Following Prophylactic Levetiracetam Administration Prior To Traumatic Brain Injury, Omeima Mutwali

Dissertations, Theses, and Capstone Projects

Traumatic brain injury (TBI) symptom prevention and remediation is an important area of research that would benefit vulnerable groups, including active-duty and veteran soldiers. These patients can sustain penetrative forces in fields of combat or in training, which result in focal lesions that trigger inflammatory and degenerative processes in the brain. Both primary and secondary injuries are associated with changes to cognition, behavior and affective state. This disease poses increased risk of epileptogenesis, as well. Given these outcomes, prior research has evaluated levetiracetam (LEV) as a prophylactic treatment for seizures, cognitive deficits and negative emotionality. LEV acts as a presynaptic …

Bayesian Spatiotemporal Model For Counterfactual Estimation In Socioeconomic Studies, Duwani W. Gonzalez

Statistical Science Theses and Dissertations

Impact evaluations of regional development programs often require estimating counterfactual outcomes for a small number of treated regions using survey-based areal data. In practice, evaluators typically rely on two-group quasi-experimental methods such as propensity score matching (PSM) and Difference-in-Differences (DiD). These approaches perform poorly when only a few regions receive treatment, and when the set of observed covariates is limited or only partially relevant. Moreover, they typically do not explicitly exploit the spatial and temporal dependence present in survey-based areal data such as in ACS (American Community Survey). This dissertation develops a family of Bayesian spatial predictive models for directly …

The Impatience Of Winning: An Analysis Of Time Discounting, Predictive Modeling, And The Nba Draft, Alec R. Plante

Business and Economics Honors Papers

This paper examines whether NBA draft decisions can be better explained by incorporating non-geometric time discounting into a model of general manager decision making. Using a dataset of 285 NBA draft prospects over a 12-year period, the impact of college statistics on Value Over Replacement Player (VORP) is determined, and these impact values are then used to create a “predicted” VORP for the first 4 seasons of each player’s career: a projection of what a general manager might think of a prospect’s future value given their college statistics. Following this, geometric and hyperbolic time discounting models are applied to estimate …

Do Dreams Reflect Our Culture? A Statistical Analysis On Dream Narratives, Michal Kuderski

Honors Capstones

Dreams are often viewed as personal experiences, but they may also reflect cultural influences. This project investigates whether dream content varies across cultures by analyzing written dream reports from American, Japanese, and Peruvian college students using data from DreamBank.net. The study applies text analysis techniques to identify common themes and compares language patterns, including the use of ‘I’ and 'We,' to examine differences in self-focus. Statistical methods for count data are used to evaluate these patterns, along with resampling to address differences in sample size. Preliminary findings suggest that both dream themes and language use may vary by cultural …

Base Running: A Lost Art In Baseball, Ethan York

Departmental Honors & Graduate Capstone Projects

In an era of baseball dominated by home runs and launch angles, the subtle art of baserunning is often overlooked, despite its measurable impact on winning games. Baserunning Runs (BsR) addresses this gap by quantifying the number of runs a player contributes through performance on the basepaths, capturing value beyond traditional metrics like stolen bases. This study constructs multiple regression models that predict BsR for Major League Baseball (MLB) players based on baserunning-related statistics. The primary objective is to examine the association between BsR and key predictors, including stolen bases (SB), extra bases taken (EB), and sprint speed (SS), while …

Empirical Comparisons Of Partial Dimension Reduction Algorithms In High-Dimensional Regression, Nathan Greenfield

Master's Theses

In high-dimensional regression problems, dimension reduction methods are often used to address the challenges of multicollinearity and estimation instability. Partial dimension reduction extends these ideas by applying dimension reduction to a subset of the predictors, while the remaining predictors are modeled without compression. This approach is particularly useful when it is important to retain variability and interpretability in certain predictors.

This thesis investigates the empirical performance of partial dimension reduction algorithms and introduces a novel algorithm, Iterative Partial Residual (IPR). Two algorithms are considered: a baseline algorithm, Marginal Residual (MR), and the proposed IPR method. Their predictive performance is evaluated …

Comparative Machine Learning Models For Disease Risk Prediction, Mercy Mawusi Agbley

Theses, Dissertations and Capstones

Accurate prediction of disease outcomes is crucial for improving clinical decision-making and enabling early intervention. This study compares the performance of various statistical and machine learning models for clinical risk prediction using two healthcare datasets: diabetic retinopathy and heart disease. The models assessed include Logistic Regression, LASSO, k-Nearest Neighbors (KNN), Support Vector Machines (SVM), Neural Networks, Random Forests, Gradient Boosting Machines (GBM), and a stacked ensemble model. Prior to modeling, datasets were split into train and test sets. Standardization was applied to numeric features whilst categorical features were one-hot encoded. These transformations were later applied to the test set. Principal …

Interpretable Sample Uncertainty Measures For Ranked-Choice Election Polls, Jason Liang

CMC Senior Theses

Polling results from traditional single-choice plurality elections are readily interpretable. Simple frequentist population parameters are estimated, including each candidate’s total support and the size of the front runner's lead. If the point estimate for the size of the front runner's lead exceeds the margin of error of the lead, we can conclude that the poll shows a statistically significant front runner. However, the interpretability of these population statistics disappears when applied to ranked-choice voting elections. Because ballots rank multiple candidates and candidates are eliminated in rounds, simple population-wide parameters are not well-defined. In RCV elections, a candidate’s ability to win …

Bayesian Analysis Of Nominal Outcomes With Missing Values Using Multinomial And Multivariate Multinomial Probit Models, Suwash Silwal

Dissertations, Master's Theses and Master's Reports

Nominal outcomes frequently arise in health sciences, transportation, economics, market research, and related fields. These data often contain missing values, while longitudinal and panel studies generate multiple correlated nominal responses. Bayesian estimation of multinomial probit (MNP) and multivariate multinomial probit (MMNP) models provides a flexible framework for analyzing such data but remains computationally challenging due to high-dimensional likelihood integration, restrictive covariance identification constraints, and poor mixing of Markov chain Monte Carlo (MCMC) algorithms, particularly in the presence of missing data. This dissertation develops parameter-expanded data augmentation (PX-DA) methods for MNP and MMNP models with missing nominal outcomes by incorporating parameter …

Modeling Housing Prices: Which Features Matter Most?, Alex Ruvolo

Williams Honors College, Honors Research Projects

This paper attempts to find the biggest factors and traits that influence the cost of housing. This will include the lot size, type of street, utilities, neighborhood, year built, heating, electrical, yard size, number of different rooms, age, condition, and others. I will attempt to answer the question of whether the prices of houses have changed within the last 5 to 10 years, and obviously this is an easy question to answer. However, the bigger question beyond this is are the main factors affecting housing prices all important in explaining this relationship? Is one factor more important than the rest …

Hybrid Patchtst And Physics-Based Framework For Predicting Lithium-Ion Battery State Of Health, Pavan Ravuri

Honors Undergraduate Theses

Accurately forecasting the state of health of lithium-ion batteries is critical for improving performance, reliability and lifetime in energy storage applications. Battery capacity degrades into a nonlinear pattern over cycling due to electrochemical processes where neither purely data driven nor physics-based models can capture alone. This study looks at a hybrid framework combining that PatchTST patch-based transformer architecture with physics-based features derived from the solid electrolyte interphase and pseudo two-dimensional models. Physics inspired proxy features were computed from cycling data and concatenated with electrochemical measurements as added input channels before patch segmentation. There are five model configurations that were evaluated …

Machine Learning-Based Spatio-Temporal Modeling Of Climate Dynamics And Desertification In The Sahara–Sahel Region, Stephen M. Tivenan

Theses and Dissertations

Arid climate classifications are threshold-dependent and easily interpretable mappings that are widely used in ecological, agricultural, and climate-related studies. These classifications inform scientific understanding, support policy and land management decisions, and provide an intuitive summary of environmental conditions. Despite their usefulness, traditional arid climate classifications often fail to quantify uncertainty, incorporate spatial context, or account for complex relationships among relevant environmental variables. Existing approaches to uncertainty assessment have largely relied on comparing classifications across multiple datasets or alternative formulas, but these methods generally overlook important spatial dependence and latent structure in the data.

This dissertation develops three machine learning-based statistical …

Statistical Quality Control: A Bayesian Framework, Jakia Jaber Tunal

College of Graduate Studies: Theses & Dissertations

In many industries, it is important to assess whether a machine or system is operating within acceptable limits or has gone out of control. This project applies Bayesian statistics to monitor a process over time and detect changes in its behavior. First, initial data are collected to understand the system’s typical performance and to form a starting prior distribution. As new observations arrive over time, the prior is updated through Bayesian inference, combining past information with incoming data. This iterative updating creates a continuous monitoring framework that adapts as more evidence becomes available. When the updated results suggest that the …

A Comparative Evaluation Of Data Imbalance Handling Techniques In Machine Learning Models For One-Year Mortality Prediction In Liver Cirrhosis, Sumiya Hasan Trisha

UNF Graduate Theses and Dissertations

Liver cirrhosis is associated with substantial morbidity and mortality, making one-year mortality prediction a clinically relevant problem. Using a liver cirrhosis dataset as the motivating application, this thesis evaluates five machine learning classifiers—Logistic Regression, Random Forest, XGBoost, LightGBM, and CatBoost—under five class-imbalance handling strategies: Baseline learning, Random Oversampling, SMOTE-NC, ADASYN, and Cost-Sensitive Learning. Hyperparameter tuning was conducted using randomized search, and predictive performance was assessed over 200 iterations of Monte Carlo Cross-Validation using Accuracy, Precision, Recall, Fl-score, and ROC-AUC.

The results suggest that imbalance-handling strategies can materially affect predictive performance, particularly recall. Because the outcome of interest is death within …

Variance Shrinkage In Dunnett-Type Multiple Comparisons With Missing Data, Md Habibullah

UNF Graduate Theses and Dissertations

Dunnett’s procedure is widely used for comparing multiple treatments with a control, but its application becomes challenging in the presence of missing data and multiple com- parisons. An improved Dunnett-type procedure addresses this by using multiple imputation under Rubin’s framework and constructing unified confidence intervals based on a multi- variate t distribution, allowing valid simultaneous inference while controlling the family-wise error rate (FWER). This work further extends the method by incorporating shrinkage-based variance estimation. Specifically, individual group variances are shrunk toward a common value to improve stability. This approach is particularly effective when group variances are similar or moderately different, …

Ridit-Based Adaptive Allocation In Two-Stage Clinical Trials With Binary Outcomes, Dewan Fahim

UNF Graduate Theses and Dissertations

This study explores better ways to assign patients to treatments in clinical trials with binary outcomes, such as success or failure. Adaptive methods are used to learn from early results and adjust treatment assignments during the trial, helping more patients receive better performing treatments while maintaining reliable conclusions. We focus on trials comparing multiple treatments using a two-stage design. In the first stage, several treatments are tested to identify the most promising one; in the second stage, that treatment is compared with a control. Unlike traditional equal assignment, we use adaptive allocation in the second stage to make better use …

A Comparative Study Of Classification Methods For Healthcare Analytics, Xueting Zhao

UNF Graduate Theses and Dissertations

This thesis presents a comparative study of logistic regression, Linear Discriminant Analy- sis (LDA), and Quadratic Discriminant Analysis (QDA) for binary classification in healthcare analytics, integrating theoretical derivation, simulation, and real-data application. A facto- rial simulation study crosses the covariance structure (equal vs. unequal), predictor correla- tion (ρ ∈ {0, 0.5, 0.9}), dimensionality (p ∈ {2, 5, 10}) and sample size (n ∈ {50, 100, 200}) across 54 scenarios with 1,000 Monte Carlo replicates each. Three main findings emerge. Logistic regression and LDA are nearly interchangeable when the assumption of equal-covariance holds. QDA achieves substantially better discrimi- nation when class-specific …

Two-Stage Response-Adaptive Randomization Designs For Multi-Arm Trials With Normal Outcome, Tanjin Tamanna Happy

UNF Graduate Theses and Dissertations

This study focuses on improving how clinical trials compare new treatments with a standard treatment, when the response is quantitative (normally distributed). In a common two-stage design, several new treatments are first evaluated, and the best-performing one is selected if it appears better than the standard. In the second stage, this selected treatment is compared again with the standard using additional data to confirm its effectiveness. This approach is known to be efficient in terms of accuracy and sample size savings. We extend this design by introducing an adaptive method for assigning patients to treatments in the second stage. Instead …

An Integrated Data-Driven Framework For Arctic Shipping: Analyzing Vessel Speed, Environmental And Ecological Factors Through Innovative Statistical Spatio-Temporal Methods, Inverse Optimization And Machine Learning, Mauli Pant

Theses and Dissertations

This dissertation develops an integrated data-driven framework to analyze vessel navigation and ecological risk in the United States Arctic from 2010 to 2019. As environmental change and maritime activity increase in the region, understanding how vessels respond to dynamic conditions and how those responses interact with marine ecosystems has become increasingly important. A central theme of this dissertation is the treatment of vessel speed as both an observed outcome and a decision variable reflecting trade- offs among operational, environmental, and ecological factors. The first chapter develops a predictive framework for vessel speed over ground (SOG) using Gaussian Process Boosting (GPBoost), …

A Geospatial Assessment Of Groundwater Salinization In A Multi-Aquifer System: Durango, Mexico, Juan Lopez-Sierra

Graduate Theses/Dissertations

Groundwater salinization poses a critical environmental concern for water resource sustainability in arid and semi-arid regions. This study evaluates spatial and temporal patterns of groundwater salinity across the state of Durango, Mexico, using total dissolved solids (TDS), sodium adsorption ratio (SAR), as salinity indicators and nitrate-nitrogen (NO₃–N) as an anthropogenic indicator. Groundwater quality data were obtained from (CONAGUA), a Mexican water agency. To assess salinity variations with respect to time, while minimizing interannual sampling bias, two multi-year sampling periods were selected: 2012-2013, and 2020-2021. Final datasets consisted of 122 wells for 2012–2013 and 131 wells for 2020–2021. The wells were …

Flexible Spatial Priors In Bayesian Neuroimaging: Gmrf, Nngp, And Deep Gmrf, Boyoung Hur

All Dissertations

Structural neuroimaging is essential for understanding neurological disorders such as Alzheimer’s disease, enabling accurate delineation of brain regions through image segmentation. Among various segmentation methods, multi-atlas-based approaches like label fusion have become leading techniques. In statistics, Bayesian hierarchical models for label fusion are increasingly favored for their ability to incorporate uncertainty and prior knowledge. Also, a key challenge in modeling neuroimaging data is spatial dependence among image voxels, making the choice of spatial prior critical—particularly in high-resolution settings where segmentation accuracy and computational efficiency are both essential.

This dissertation proposes fully Bayesian spatial hierarchical models that explore two flex- ible …

Spatiotemporal Modeling Of Maternal Mortality In South Carolina 2018-2023, Leah Wood, Ray Bai, Emily Mann

Senior Theses

Maternal death serves as a public health indicator due to fact that it is considered preventable with the availability of modern biomedicine, however, it persists broadly throughout the United States. Current literature outlines national trends in maternal mortality with complicating, preexisting conditions, and structural upstream factors often cited as being the largest contributors to increased risk. This study utilizes publicly available, county-level data for maternal death in addition to demographic and descriptive data in order to estimate maternal mortality rates in each of South Carolina’s 46 counties from 2018 to 2023. In order to address sparsity in the outcome variable …

Comparing Ordinary Least Squares And Quantile Regression: A Causal-Comparative Approach To Modeling Conditional Relationships, Samuel Nnorom

Electronic Theses and Dissertations

Ordinary Least Squares (OLS) regression has traditionally been the preferred quantitative method for estimating linear relationships. However, it assumes that the effect of a predictor variable remains constant across the entire outcome distribution, which can miss important insights when data are heterogeneous. Quantile Regression (QR), on the other hand, offers a more detailed analysis by focusing on the full response variable distribution, thereby revealing different relationship patterns at various quantiles within the outcome. This study compares how OLS and QR perform in modeling conditional relationships within a causal-comparative framework based on ex post facto research. Using the mortality data from …

Towards Reliable Clinical Applications Of Ai Models In Radiotherapy, Biling Wang

Statistical Science Theses and Dissertations

Over the past decade, artificial intelligence (AI), particularly through deep learning (DL) techniques, has made significant strides in fields like computer vision (CV) and natural language processing (NLP), leading to transformative advancements across numerous applications. This progress has sparked considerable enthusiasm within the medical field, where DL-related research has grown exponentially since 2015. However, despite these promising developments, the real-world deployment of DL models in healthcare remains limited, especially in safety-critical domains such as radiotherapy (RT), where reliability, safety, and sustained performance are critical. This thesis addresses three core challenges associated with the clinical application of DL models: (1) post-deployment …

Estimation Methods For Bayesian Exponential Random Graph Models Under The Horseshoe Prior., Pamela Linares

Electronic Theses and Dissertations

Networks are powerful tools for modeling the complexity of social interactions, biological systems, and information spread. A leading statistical frameworks for analyzing network data are Exponential Random Graph Models (ERGMs), which provide a principled approach to capturing structural dependencies. However, ERGMs remain challenging to estimate, especially in sparse or high-dimensional settings where models suffer from degeneracy and unstable parameter inference. This paper proposes a penalized Bayesian approach to ERGMs that utilizes the horseshoe prior, a sparsity-inducing global-local shrinkage prior. This prior offers robust regularization while preserving important signals, improving estimation by shrinking irrelevant parameters and reducing the impact of extreme …