Statistics and Probability Commons^™

Fossil-Fuels In A Decarbonized Country? Modeling The Drivers Of Icelandic Oil Sales, Inbal Armony

Environmental Studies Honors Projects

Although 100% of Iceland’s electricity comes from renewable energy sources, it still relies on fossil fuels for land transportation, marine transportation, aviation, and some industry. Understanding geographic nuances in oil use is critical to achieving Iceland’s goals of carbon neutrality by 2040. As the island has one primary urban center with two thirds of the population, information is lacking about oil use in non-Capital areas and a gap between state and municipal climate plans. Using newly available data of oil sales at the municipality-level in a Small Area Estimation model, we analyze drivers of oil sales across Icelandic municipalities. We …

Progress, Characteristics And Prospects Of Interdisciplinary Research Collaboration At Home And Abroad:Research Based On Literature Published In The Past Five Years, Yueliang Zeng, Ying Yu, Ruirui Gao

Journal of Scientific Information Research

[Purpose/significance] Interdisciplinary research collaboration is an important way to promote knowledge innovation and solve complex social problems in the new era. Reviewing the research progress of interdisciplinary research collaboration at home and abroad in the past five years can help grasp the research frontiers in this field and provide direction for future research. [Method/process] This paper adopts a systematic review method, selecting literature closely related to the theme of interdisciplinary research collaboration from inWeb of Science and CNKI from 2020 to 2024, condensing the research topic, analyzing the main research viewpoints, and summarizing the changing characteristics, then proposing future research …

Research On Academic Journal Evaluation Based On Factor Dimensionality Reduction, Data Weighting, And Journal Characteristics Orientation, Liping Yu, Jing Zhang

Journal of Scientific Information Research

[Purpose/significance] The characteristics of academic journals have important value, and current evaluation systems lack assesment for this dimension. [Method/process] This paper proposes a variable weight method for factor reduction data, firstly uses factor analysis combined with manual classification to determine the characteristics of the journal, and uses Sigmoid function to standardize the public factors to determine the characteristic journals, and then uses the variable weight function to modify the original data for the characteristic indicators of the characteristic journals based on the data of forestry journals in CNKI. Three representative methods including linear weighted aggregation, weighted TOPSIS, and factor analysis, …

A Meta-Analysis Of Corporate Innovation Intention And Its Driving Mechanisms, Yaping Hu, Xuerong Shen, Yi Zheng

Journal of Scientific Information Research

[Purpose/significance] Stimulating corporate innovation intention is a pivotal issue for promoting innovation. However, the academic discourse on its key drivers presents significantly divergent and even contradictory conclusions, leading to theoretical ambiguity and practical guidance challenges. This study aims to systematically and quantitatively integrate empirical research in this field to clarify the true effects of core driving factors and their operational boundaries. [Method/process] Adopting a meta-analysis approach, this study systematically retrieves and screens literature, ultimately analyzes 29 empirical studies and examine four potential moderating variables which include location, data type, industry, and culture. [Result/conclusion] The study finds that government support, internal …

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 …

Pyspqr: A Python Package For Density Estimation Using Deep Learning, Cameron Eddy, Reetam Majumder

Electrical Engineering and Computer Science Undergraduate Honors Theses

Splines are used for representing complex functions. In statistics, splines can be used for distributional shapes that are difficult to model by traditional parametric approaches. Ramsay (1) uses M-Spline bases to estimate continuous distributions. Semi-Parametric Quantile Regression (SPQR), developed by Xu and Reich (2), models conditional distributions where a neural network is used to estimate the basis function weights that depend on covariates. (3) implements a package for SPQR in R. We build on this by implementing a version of SPQR in Python with PyTorch. By using PyTorch, we can use more sophisticated deep learning architectures than those available in …

Multi-Population Sufficient Dimension Reduction, Xuerong Meggie Wen, Yuexiao Dong, Li Xing Zhu

Mathematics and Statistics Faculty Research & Creative Works

A novel dimension-reduction method is introduced for multi-population data. The approach conducts a joint analysis that exploits information shared across populations while accommodating population-specific effects. Unlike partial dimension reduction methods, which identify related directions across all populations, or conditional analyses conducted independently within each population, the proposed two-step procedure leverages cross-population information to enhance estimation accuracy. The methodology is demonstrated through simulations and two real-data applications.

Best Linear Unbiased Prediction Based On Optimally Selected Order Statistics Using Compound Design, Foster Nkansah

Open Access Theses & Dissertations

Any life-testing experiments with censoring generate ordered failure data, which are used to estimate model parameters of the underlying lifetime distribution. However, predicting unobserved failure times is an important goal post inference. Balakrishnan and Bhattacharya [2021] showed that joint best linear unbiased predictors (BLUPs) are more efficient than their marginal counterparts. In this thesis, we introduce a compound optimal design to compute joint BLUPs based on a suitably chosen subset of the observed data. It is shown that the linear predictors of the parameters from any location-scale distribution, based on a few optimally chosen ranks, obtained from the compound optimal …

Analyzing The Evolution Of Science: Topological Cycles And Community Detection In Knowledge Networks, Frances C. Mcconnell

Mathematics, Statistics, and Computer Science Honors Projects

How scientific knowledge grows and organizes itself is a central question in the study of science. This thesis uses tools from topology and network science to detect and characterize knowledge gaps—places in a field’s literature where related concepts do not co-occur. We develop a metric to quantify the degree of interdisciplinarity of each gap, using the community structure of the underlying network as a proxy for subfields. Across a wide range of fields, gaps reliably span multiple subfields and evolve in recognizable temporal patterns, highlighting new insights into how scientific fields are structured and their stage of development.

Understanding Delays In Emergency Department Care: A National Analysis Of Wait Times, Gregory Forsberg

Mathematics, Statistics, and Computer Science Honors Projects

Emergency department (ED) wait times remain a persistent bottleneck in the United States healthcare system, impacting patient outcomes, hospital efficiency, and equitable access to care. This study analyzes nationally representative data from the National Hospital Ambulatory Medical Care Survey (NHAMCS), a complex, multi-stage probability sample. Using survey-weighted analyses and predictive modeling, we examine the effects of patient characteristics, triage acuity, and visit timing. Results indicate that operational and system-level factors, including hospital capacity, geographic region, and temporal variation, are among the most influential predictors of ED wait times

Evaluation Of The Effectiveness Of Antihypertensive Therapy Chosen Through Maternal Hemodynamic Profile Analysis, Margaret Weimer

Poster Presentations

In treatment of emergent hypertension, the American College of Obstetricians and Gynecologists (ACOG) recommends both oral labetalol and nifedipine, with no preference shown between the two. These medications have different mechanisms of action. Studies have suggested that consideration of hemodynamic profile may be an effective way to determine medication, but have not considered the Rule of 55, a simple calculation to predict hemodynamic profile.

Goals: 1. Determine if tailoring the antihypertensive to patient’s hemodynamic profile is associated with better control of blood pressure. 2. Determine if the Rule of 55 is an accurate method to determine hemodynamic profile.

Perseus Interim Evaluation Report, Data Innovation Project, University Of Southern Maine

Publications

The Data Innovation Project (DIP) from the University of Southern Maine’s Catherine Cutler Institute serves as the external evaluator for the USDA NIFA SAS-funded Promoting Economic Resilience and Sustainability of the Eastern U.S. Forests (PERSEUS) project. This project is a joint effort between Purdue University, the University of Georgia, and the University of Maine. The DIP supports the project in assessing its progress, implementation, and emerging impacts through data collection from project participants and systematic document review (i.e. annual USDA reports, personnel lists, and meeting notes). This includes synthesizing quantitative and qualitative data and incorporating these metrics into independent evaluation …

Survival Patterns Among Adult And Pediatric Bone Cancer Patients, Ethan Estes

Mathematical Sciences Undergraduate Honors Theses

Recently noted, Huang et al. (2023), machine learning (ML) models, while offering great advantages over traditional statistical predictive modeling methods, are less explored in the analysis of survival and other similar time-to-event predictive data modeling. ML methods such as neural networks offer a great deal of promise but need to be further explored to investigate their comparative power in predicting survival outcomes. Focusing specifically on survival analysis in adult and pediatric bone cancer patients, traditional methods, like shown in Emmert-Streib and Dehmer (2019), will be shown with machine learning models using methods in Hothorn, Hornik, and Zeileis (2006). In this …

Explainability In Deep Learning For Density Regression, Dalton James Oxford

Graduate Theses and Dissertations

Classical statistical methods focus on explainability and inferential power. Machine learning and deep learning can handle non-linear, high-dimensional data better than traditional methods. In modeling, a clear understanding and interpretation are essential to decision-making. Recent work in quantile regression and extreme modeling has begun to use deep learning due to its performance on high-dimensional, non-linear data. Semi-Parametric Quantile Regression (SPQR) is a nonparametric spline-based approach to quantile regression that estimates the conditional PDF and CDF of the response. Semi-Parametric Quantile Regression for Extremes (SPQRx) is a recent extension of SPQR that provides two features: out-of-sample estimation and accurate extreme-tailed estimation. …

Factors Affecting Pedestrian-Vehicle Conflicts: An Empirical Analysis, Christo D. Jamo

Honors Scholar Theses

The number of pedestrian deaths increased by 78% between 2009 and 2023, while other motor vehicle crash deaths increased by 13% in the same period [1]. To identify potential pedestrian safety measures, this study analyzed the effects of location-based demographics, pedestrian phasing type, and other physical infrastructure and behavior variables on the probability of pedestrian-vehicle conflicts at signalized intersections, which is a surrogate measure of crash risk. Data were collected from 55 intersections in Connecticut, and the pedestrian-vehicle interactions were classified by severity based on the Swedish Traffic Conflict Technique: undisturbed passage, potential conflict, minor conflict, or serious conflict. Because …

A Novel Approach To Creativity Assessment: Forced Pairwise Ranking With Large Language Models, Phillip R. Gregory Jr

Master's Theses

Assessing creativity at scale remains a persistent challenge in cognitive science, as human raters are costly, slow, and often inconsistent in their judgments. This thesis introduced a novel framework for automated scientific creativity assessment using forced pairwise ranking, in which fine-tuned large language models compared response pairs and determined which was more creative. Five empirical studies were conducted using Llama-2-7B and Llama-2-13B models adapted via LoRA fine-tuning and benchmarked against human scored responses from the Scientific Creative Thinking Test. A regression baseline achieved Pearson �� = .74 on the test set, matching the human inter-rater ceiling reported in the literature. …

Pattern Dynamics And Stochasticity Of Brain Rhythms And Spike Trains In A Tauopathy Mouse Model Of Alzheimer’S Disease, Clarissa M. Hoffman

Dissertations and Theses (Open Access)

Systems neuroscience posits that every aspect of perceived physical reality, every aspect of animal and human behavior, and every cognitive phenomenon emerges from patterns of neuronal activity. While most researchers embrace this idea, there are major difficulties in describing and analyzing these complex neuronal dynamics—spike flows produced by cells ensembles, synchronized extracellular field oscillations, and other patterns—which limits our understanding of how the activity of individual neurons and the whole-animal cognition and behavior might be connected. In particular, we lack the approaches and even the semantics for connecting the individual cell outputs and the integrated results of their activity. Current …

A New Approach To Generate Combinatorial Patterns In Logical Analysis Of Data And Its Application To Predict College Retention, Salihah Ahmed E. Jaafari

Theses and Dissertations

Student retention and degree completion remain central challenges for higher-education institutions, with significant implications for student success, institutional effectiveness, and public accountability. While advances in predictive analytics have enabled earlier identification of students at risk of withdrawal, many commonly used machine learning approaches suffer from limited interpretability, constraining their practical usefulness for advising, intervention, and policy decision making. This dissertation addresses the problem of predicting student persistence by developing and evaluating optimization based, interpretable classification models within the Logical Analysis of Data (LAD) framework. Building on existing LAD formulations, this research introduces two novel pattern generation models, the Best Term …

Log Anomaly Detection With Parameter-Efficient Tiny Language Models: From Centralized Fine-Tuning To Privacy-Preserving Federated Learning, Isaiah Thompson Thompson Ocansey

Open Access Theses & Dissertations

System logs are a primary source of information for identifying faults, misconfigurations, and security incidents in computing infrastructure. As these logs grow in volume and complexity, manual inspection becomes impractical, and automated detection methods are needed. This thesis investigates how compact language models can be applied to the task of log anomaly detection under two operational conditions: when log data is centralized and when it is located at different data sites.

In the first part, we propose LogTinyLLM, which applies parameter-efficient fine-tuning through Low-Rank Adaptation and adapter-based techniques to adapt tiny language models for identifying contextual anomalies in log sequences. …

An Ordinal Categorical Data Analysis Using The Stereotype Model Within A Bayesian Framework With A Logistic-Normal Prior, Nathaniel A. Sakyi

Open Access Theses & Dissertations

Ordinal categorical data are pervasive in applied research, yet their analysis is often compromised by modeling strategies that implicitly assume equidistant category spacing or impose restrictive structural constraints. Metric regression models and latent-variable threshold approaches routinely misrepresent ordinal information, leading to biased inference, distorted uncertainty quantification, and loss of structural insight. This paper develops a Bayesian framework for ordinal regression that explicitly accommodates \textbf{unequal spacing among ordered response categories} while preserving ordinal structure and interpretability.

The proposed approach builds on the stereotype regression model, which embeds ordinal categories into a latent one-dimensional continuum through estimable score parameters. While the stereotype …