Statistics and Probability Commons^™

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 …

Using Camera-Based Unmarked Spatial Capture-Recapture Modeling To Estimate Reintroduced Elk (Cervus Canadensis) Population Parameters And Distribution In Southeastern Kentucky, Claire Marie Muia

Theses and Dissertations--Forestry and Natural Resources

Estimation of population parameters is important for wildlife management decisions. Elk reintroduced to southeastern Kentucky experienced early irruptive population growth and are currently monitored using a statewide harvest-based statistical population reconstruction model (SPR) across the Kentucky Elk Restoration Zone (KERZ). Because the SPR model is spatially coarse and difficult to scale to the smaller management units comprising the KERZ, we conducted a spatially explicit capture-recapture study using a clustered camera-trapping array deployed for 10 weeks from June–August 2024 to estimate elk population parameters within Management Unit 4. Due to a lack of resights of GPS-marked elk, population parameters were estimated …

On Bayesian Empirical Likelihood-Based Method For Complex Survey Data With Application To Non-Probability Sampling, Md Hasibur Rahman

Department of Statistics: Dissertations, Theses, and Student Research

This thesis develops a Bayesian empirical likelihood (BEL) framework for inference under complex survey designs and extends it to non-probability sampling. Parametric likelihood based methods are difficult to apply to complex survey data because the likelihood is rarely available in closed form. EL provides a flexible alternative by replacing the parametric likelihood with an empirical likelihood constructed from moment conditions. The proposed method first integrates empirical likelihood constraints with survey design features then extends BEL to non-probability sampling through selection models and design consistent restrictions. Posterior inference is carried out using a Metropolis–Hastings MCMC algorithm. A real-data analysis further illustrates …

Latent Variable Dyadic Regression Models For Predicting Over/Under Bets In Sports Betting, Alexcia Trejo

Graduate Theses and Dissertations

This thesis explores the use of latent factor models to uncover hidden structures in pair wise outcomes derived from Over/Under betting markets in sports betting. Specifically, we implement and evaluate the Eigen model, a latent space model that represents dyadic data using node-specific vectors whose inner product govern edge probabilities. By modeling relationships between teams as adjacency matrices of binary outcomes, we investigate the extent to which the Eigen model captures both homophily, the tendency of similar teams to yield consistent betting results, and stochastic equivalence, where different teams exhibit indistinguishable patterns of Over/Under outcomes. A Bayesian formulation of the …

Bayesian Merged Utilization Of Grappa And Sense (Bmugs) For In-Plane Accelerated Reconstruction Increases Fmri Detection Power, Chase J. Sakitis, Daniel B. Rowe

Mathematical and Statistical Science Faculty Research and Publications

In fMRI, capturing brain activity during a task is dependent on how quickly the k-space arrays for each volume image are obtained. Acquiring the full k-space arrays can take a considerable amount of time. Under-sampling k-space reduces the acquisition time, but results in aliased, or “folded,” images after applying the inverse Fourier transform (IFT). GeneRalized Autocalibrating Partial Parallel Acquisition (GRAPPA) and SENSitivity Encoding (SENSE) are parallel imaging techniques that yield reconstructed images from subsampled arrays of k-space. With GRAPPA operating in the spatial frequency domain and SENSE in image space, these techniques have been separate but can …

A Bayesian Complex-Valued Latent Variable Model Applied To Functional Magnetic Resonance Imaging, Chase J. Sakitis, D. Andrew Brown, Daniel B. Rowe

Mathematical and Statistical Science Faculty Research and Publications

In linear regression, the coefficients are simple to estimate using the least squares method with a known design matrix for the observed measurements. However, real-world applications may encounter complications such as an unknown design matrix and complex-valued parameters. The design matrix can be estimated from prior information but can potentially cause an inverse problem when multiplying by the transpose as it is generally ill-conditioned. This can be combat by adding regularizers to the model but does not always mitigate the issues. Here, we propose our Bayesian approach to a complex-valued latent variable linear model with an application to functional magnetic …

Using Neighborhood Information To Improve Fiber Direction Estimation From Neuroimaging Data, Anjan Mandal

UNLV Theses, Dissertations, Professional Papers, and Capstones

Accurately estimating neuronal fiber directions is crucial in neuroimaging analysis. The Ball-and-Stick Model (BSM), introduced by Behrens et al. (2003), and widely used in software tools like FSL, remains one of the most popular models for this purpose. BSM analyzes each voxel individually, based solely on its signal information.In this dissertation, we propose modifications to BSM, that incorporate signal information from neighboring voxels in the estimation process, potentially improving the accuracy of the estimates. Additionally, within the estimation process, we introduce two novel proposal distributions to enhance the efficiency of the Markov chain Monte Carlo sampling procedure on the simplex …

Bayesian Methods In Analyzing The Diagnostic Accuracy\\ For Ordinal Ratings, Yun Yang

Theses and Dissertations

This dissertation focuses on ordinal classification ratings, which are commonly used in medical practice to assess the severity of a disease or condition. For example, a group of radiologists rate a set of mammograms and assign BI-RADS (Breast Imaging Reporting Data System) score for each mammogram. A Bayesian probit hierarchical model is first proposed to analyze this type of data. It links the ordinal ratings with both rater diagnostic skills and patient latent disease severity. Each rater diagnostic skills are quantified with two parameters, diagnostic bias and diagnostic magnifier. Patient latent disease severity is assumed to follow a different normal …

Disentangling Cyclic Causality: An Instance-Based Framework For Causal Discovery, Chase A. Yakaboski

Dartmouth College Ph.D Dissertations

Correlation does not imply causation" is one of the fundamental principles taught in science, emphasizing that associations between variables do not necessarily indicate causality. Yet, over the past three decades, extensive research has begun to challenge this perspective by developing sophisticated methods to differentiate causal from correlative relationships. This research suggests that correlations often involve a blend of confounded and causal interactions, which, given certain assumptions, can be disentangled to uncover actionable insights and deepen our understanding of physical, biological, and societal systems.

Accurately discovering causal relationships from data amidst cyclic dynamics remains a challenging open problem in causality research. …

Comparative Analysis Of Teacher Effects Parameters In Models Used For Assessing School Effectiveness: Value-Added Models & Persistence, Merlin J. Kamgue

Graduate Theses and Dissertations

Longitudinal measures for students have become increasingly popular to estimate the effects of individual teachers and schools. Value-added models are one of the approaches using longitudinal data to evaluate teachers and schools. In the value-added model (VAM) literature, many statistical approaches have been developed and used to estimate teacher or school effects on student learning. This study opted to use a Bayesian multivariate model for evaluating teacher effects. The generalized persistence models can handle longitudinal data, not vertically scaled, allowing for a below-par teacher’s effects correlation across test administrations. This study first generated longitudinal students’ test score data and used …

A New Method To Determine The Posterior Distribution Of Coefficient Alpha, John Mart V. Delosreyes

Psychology Theses & Dissertations

There is a focus within the behavioral/social sciences on non-physical, psychological constructs (i.e., constructs). These constructs are indirectly measured using measurement instruments that consist of questions that capture the manifestations of these constructs. The indirect nature of measuring constructs results in a need of ensuring that measurement instruments are reliable. The most popular statistic used to estimate reliability is coefficient alpha as it is easy to compute and has properties that make it desirable to use. Coefficient alpha’s popularity has resulted in a wide breadth of research into its qualities. Notably, research about coefficient alpha’s distribution has led to developments …

Penalized Bayesian Exponential Random Graph Models., Vicki Modisette

Electronic Theses and Dissertations

Networks have the critical ability to represent the complex interconnectedness of social relationships, biological processes, and the spread of diseases and information. Exponential random graph models (ERGM) are one of the popular statistical methods for analyzing network data. ERGM, however, struggle with computational challenges and degeneracy issues, further exacerbated by their inability to handle high-dimensional network data. Bayesian techniques provide a promising avenue to overcome these two problems. This paper considers penalized Bayesian exponential random graph models with adaptive lasso and adaptive ridge penalties to perform variable selection and reduce multicollinearity on a variety of networks. The experimental results demonstrate …

Spatially Adaptive Estimation Of Spectrum, Yi Xie

Open Access Theses & Dissertations

A time series may be analyzed either in the time or in the frequency domain. When working in the frequency domain, the main objective is to estimate the underlying spectrum. Various approaches have been proposed to this end, but most are based on smoothing the periodogram using a single smoothing parameter across all Fourier frequencies. Such a global smoothing parameter may result in a biased estimate. To improve the estimation, in this paper, we smooth the log periodogram by placing a dynamic shrinkage prior, such that varying degrees of smoothing may be applied to different regions of the Fourier frequencies, …

Bayesian Dependence Structure Analysis For Ordinal Data, Yang He

Theses and Dissertations

This dissertation explores different methods to study the dependence structure among many ordinal variables under the Bayesian framework.

Chapter 1 introduces ordinal data analysis methods, and the related literature works are briefly reviewed. An outline of the dissertation is put forward.

In Chapter 2, Gaussian copula graphical models with different priors of graphical Lasso, adaptive graphical Lasso, and spike-and-slab Lasso on the precision matrix are assessed and compared. The proposed models are well illustrated via simulations and a real ordinal survey data analysis.

In Chapter 3, adaptive spike-and-slab Lasso prior is proposed as an extension of Chapter 2. The developed …

Self-Learning Algorithms For Intrusion Detection And Prevention Systems (Idps), Juan E. Nunez, Roger W. Tchegui Donfack, Rohit Rohit, Hayley Horn

SMU Data Science Review

Today, there is an increased risk to data privacy and information security due to cyberattacks that compromise data reliability and accessibility. New machine learning models are needed to detect and prevent these cyberattacks. One application of these models is cybersecurity threat detection and prevention systems that can create a baseline of a network's traffic patterns to detect anomalies without needing pre-labeled data; thus, enabling the identification of abnormal network events as threats. This research explored algorithms that can help automate anomaly detection on an enterprise network using Canadian Institute for Cybersecurity data. This study demonstrates that Neural Networks with Bayesian …

Model-Based Imputation Of Below Detection Limit Missing Data And Group Selection In Bayesian Group Index Regression, Matthew Carli

Theses and Dissertations

Investigations into the association between chemical exposure and health outcomes are increasingly focused on the role of chemical mixtures, as opposed to individual chemicals. The analysis of chemical mixture data required the development of novel statistical methods, one of these being Bayesian group index regression. A statistical challenge common to all chemical mixture analyses is the ubiquitous presence of below detection limit (BDL) data. We propose an extension of Bayesian group index regression that treats both regression effects and missing BDL observations as parameters in a model estimated through a Markov Chain Monte Carlo algorithm that we refer to as …

Bayesian Methods For Graphical Models With Neighborhood Selection., Sagnik Bhadury

Electronic Theses and Dissertations

Graphical models determine associations between variables through the notion of conditional independence. Gaussian graphical models are a widely used class of such models, where the relationships are formalized by non-null entries of the precision matrix. However, in high-dimensional cases, covariance estimates are typically unstable. Moreover, it is natural to expect only a few significant associations to be present in many realistic applications. This necessitates the injection of sparsity techniques into the estimation method. Classical frequentist methods, like GLASSO, use penalization techniques for this purpose. Fully Bayesian methods, on the contrary, are slow because they require iteratively sampling over a quadratic …

Power Approximations For Generalized Linear Mixed Models In R Using Steep Priors On Variance Components, Sydney Geisler

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

When designing an experiment, researchers often want to know how likely they are to detect statistically significant effects in the resulting data, i.e., they want to estimate their statistical power. The probability distribution method is a flexible way to do this, and it is currently implemented in the statistical software package SAS. This method requires a hypothetical data set (showing the magnitude of hypothesized effects) and constant values of variance components, which are critical elements of the statistical models used. The statistical software package R is increasingly popular, but the probability distribution method has not yet been implemented in R, …

Dataset Evaluation For Data Trading Using Expected Loss And Homomorphic Encryption, Minsung Joo

Senior Honors Papers / Undergraduate Theses

Supervised machine learning suffers from the ``garbage-in garbage-out" phenomenon where the performance of a model is limited by the quality of the data. While a myriad of data is collected every second, there is no general rigorous method of evaluating the quality of a given dataset. This hinders fair pricing of data in scenarios where a buyer may look to buy data for use with machine learning. In this work, I propose using the expected loss corresponding to a dataset as a measure of its quality, relying on Bayesian methods for uncertainty quantification. Furthermore, I present a secure multi-party computation …

A Copula Model Approach To Identify The Differential Gene Expression, Prasansha Liyanaarachchi

Mathematics & Statistics Theses & Dissertations

Deoxyribonucleic acid, more commonly known as DNA, is a complex double helix-shaped molecule present in all living organisms and hosts thousands of genes. However, only a few genes exhibit differential expression and play a vital role in a particular disease such as breast cancer. Microarray technology is one of the modern technologies developed to study these gene expressions. There are two major microarray technologies available for expression analysis: Spotted cDNA array and oligonucleotide array. The focus of our research is the statistical analysis of data that arises from the spotted cDNA microarray. Numerous models have been proposed in the literature …

A Comparison Of Bayesian Spatial Models For Hiv Mapping In South Africa, Kassahun Abere Ayalew, Samuel Manda, Bo Cai Ph.D.

Faculty Publications

Despite making significant progress in tackling its HIV epidemic, South Africa, with 7.7 million people living with HIV, still has the biggest HIV epidemic in the world. The Government, in collaboration with developmental partners and agencies, has been strengthening its responses to the HIV epidemic to better target the delivery of HIV care, treatment strategies and prevention services. Population-based household HIV surveys have, over time, contributed to the country’s efforts in monitoring and understanding the magnitude and heterogeneity of the HIV epidemic. Local-level monitoring of progress made against HIV and AIDS is increasingly needed for decision making. Previous studies have …

Bayesian Calibration Of The Icrp Zirconium Biokinetic Model And Use Of Canned Priors For The Evaluation Of Bioassay, Thomas Raymond Labone

Theses and Dissertations

The International Commission on Radiological Protection (ICRP) publishes biokinetic models that relate measurements of radioactive material in the body and excreta (bioassay) to the amount of the material taken into the body (intake). Given the intake and the biokinetic model, radiation dose to organs and tissues can be calculated. The ICRP approximates the biokinetics of radioactive materials in the body with compartmental models expressed mathematically as a system of ordinary differential equations, for which they provide point estimates of the rate constants. Inaccurate estimates of intake and radiation dose can result in cases where the biokinetics of an individual differ …

Optimal Transport Driven Bayesian Inversion With Application To Signal Processing, Elijah F. Perez

Mathematics & Statistics ETDs

This paper will outline a Debiased Sinkhorn Divergence driven Bayesian inversion framework. Conventionally, a Gaussian Driven Bayesian framework is used when performing Bayesian inversion. A major issue with this Gaussian framework is that the Gaussian likelihood, driven by the L2 norm, is not affected by phase shift in a given signal. This issue has been addressed in [1] using a Wasserstein framework. However, the Wasserstein framework still has an issue because it assumes statistical independence when multidimensional signals are analyzed. This assumption of statistical independence cannot always be made when analyzing signals where multiple detectors are recording one event, say …

Bayesian Nonparametric Model For Functional Data Analysis, Tahmidul Islam

Theses and Dissertations

Functional data analysis (FDA) experienced a burst of growth after Ramsay and Silverman published their textbook in 1997. Functional data analysis interests researchers because of the challenges it adds to well-established multivariate analysis. Unlike finite dimensional random vectors, we visualize infinite dimensional random functions; for example, curves, images, brain scans, etc. A vast amount of literature have been dedicated to developing models for functional data. The ideas are mostly based on basis function representations and kernel-based nonparametric methods. In this dissertation, we propose a Bayesian treatment of nonparametric functional data analysis by introducing a Gaussian process (GP) over the space …

Computing For Numeracy: How Safe Is Your Covid-19 Social Bubble?, Charles Connor

Numeracy

The COVID-19 pandemic has led many people to form social bubbles. These social bubbles are small groups of people who interact with one another but restrict interactions with the outside world. The assumption in forming social bubbles is that risk of infection and severe outcomes, like hospitalization, are reduced. How effective are social bubbles? A Bayesian event tree is developed to calculate the probabilities of specific outcomes, like hospitalization, using example rates of infection in the greater community and example prior functions describing the effectiveness of isolation by members of the social bubble. The probabilities are solved for two contrasting …

Parametric, Nonparametric, And Semiparametric Linear Regression In Classical And Bayesian Statistical Quality Control, Chelsea L. Jones

Theses and Dissertations

Statistical process control (SPC) is used in many fields to understand and monitor desired processes, such as manufacturing, public health, and network traffic. SPC is categorized into two phases; in Phase I historical data is used to inform parameter estimates for a statistical model and Phase II implements this statistical model to monitor a live ongoing process. Within both phases, profile monitoring is a method to understand the functional relationship between response and explanatory variables by estimating and tracking its parameters. In profile monitoring, control charts are often used as graphical tools to visually observe process behaviors. We construct a …

Modified-Half-Normal Distribution And Different Methods To Estimate Average Treatment Effect., Jingchao Sun

Electronic Theses and Dissertations

This dissertation consists of three projects related to Modified-Half-Normal distribution and causal inference. In my first project, a new distribution called Modified-Half-Normal distribution was introduced. I explored a few of its distributional properties, the procedures for generating random samples based on Bayesian approaches, and the parameter estimation based on the method of moments. The second project deals with the problem of selection bias of average treatment effect (ATE) if we use the observational data. I combined the propensity score based inverse probability of treatment weighting (IPTW) method and the directed acyclic graph (DAG) to solve this problem. The third project …

Bayesian Analysis Of Extended Cox Model With Time-Varying Covariates Using Bootstrap Prior, Oyebayo R. Olaniran, Mohd Asrul A. Abdullah

Journal of Modern Applied Statistical Methods

A new Bayesian estimation procedure for extended cox model with time varying covariate was presented. The prior was determined using bootstrapping technique within the framework of parametric empirical Bayes. The efficiency of the proposed method was observed using Monte Carlo simulation of extended Cox model with time varying covariates under varying scenarios. Validity of the proposed method was also ascertained using real life data set of Stanford heart transplant. Comparison of the proposed method with its competitor established appreciable supremacy of the method.

Methods Of Uncertainty Quantification For Physical Parameters, Kellin Rumsey

Mathematics & Statistics ETDs

Uncertainty Quantification (UQ) is an umbrella term referring to a broad class of methods which typically involve the combination of computational modeling, experimental data and expert knowledge to study a physical system. A parameter, in the usual statistical sense, is said to be physical if it has a meaningful interpretation with respect to the physical system. Physical parameters can be viewed as inherent properties of a physical process and have a corresponding true value. Statistical inference for physical parameters is a challenging problem in UQ due to the inadequacy of the computer model. In this thesis, we provide a comprehensive …

Bayesian Zero-Inflated Model For Ordinal Data, Huizhong Yang

Theses and Dissertations

Datasets with a relatively large number of zeros is commonly seen in medical applications. Although models like Zero-inflated Poisson (ZIP) model are proposed for counts data, there is still some issues with ordinal data which have excess zeros. In this paper, we developed a Bayesian approach to accommodate the excess zero in ordinal data. Intellectual disability (ID), also known as mental retardation (MR), is a disability characterized by below-average intelligence or mental ability and a lack of the learning necessary skills for daily life. A person with intellectual disability has intellectual functioning and adaptive behaviors limitations. Intellectual disability is a …