Doctor of Philosophy, The Ohio State University, 2022, City and Regional Planning

Socio-demographics may represent constraints that shape different travel outcomes of individuals. This leads to studies with not only different findings on travel behavior, but also mixed and inconclusive conclusions on the effects of built environment on individuals' travel outcomes. There are gaps in many existing studies on the relationship between socio-demographics, built environment, and travel behavior, which need to be filled. In addition, the existing literature has not paid much attention to the varying impacts of neighborhood-built environment on travel outcomes across different socio-demographic groups. Many signs from U.S. Census Bureau and Center for Diseases Control and Prevention (CDC) indicate that the socio-demographics of the U.S. society are undergoing a process of significant changes. It is uncertain how these changes may affect travel behavior in the short term and the long term. In the face of this uncertainty, a key challenge for transportation planners and policymakers is to understand how socio-demographics affect individuals' travel outcomes and out-of-home activities. These major trends that affect future travel patterns will dramatically reshape transportation priorities and needs. This dissertation quantitatively examines the links between socio-demographic constraints, neighborhood-built environments, and travel behavior. This dissertation comprises three essays. The first essay explores gender differences in commute behavior with a focus on two-earner households. The second essay examines the links between walkability and transit use, focusing on the differences between disabled individuals and others. The third essay explores how neighborhood walkability affects older adults' walking trips, considering different household income levels. The first essay utilizes the detailed individual-level data from 2001, 2009, and 2017 National Household Travel Surveys (NHTS). The NHTS datasets provide information on travel by U.S. residen (open full item for complete abstract)

Committee: Gulsah Akar (Advisor); Zhenhua Chen (Advisor); Harvey J. Miller (Committee Member) Subjects: Transportation Planning; Urban Planning

Doctor of Philosophy, University of Akron, 2021, Civil Engineering

Adequate rebar-concrete bonding is crucial to ensure the reliable performance of reinforced concrete (RC) structures. Many factors (such as the concrete properties, concrete cover depth, transverse reinforcement, and the presence of corrosion) affect the bond behavior, and consequently the structural performance. This bond behavior is typically described by a bond stress-slip relationship, where there are two critical quantities: bond strength the maximum shear stress that bond can withstand, and peak slip the slippage at the interface when the bond strength is reached. It is understood that the bond deteriorates when corrosion is present and behaves differently under two distinct bond failure modes (i.e., splitting and pull-out). While many prior studies have focused on the influence of the aforementioned factors on the bond strength, the impact of the failure mode coupled with corrosion on the bond stress-slip relationship and structural performance have not been thoroughly investigated. This study is aimed to address this issue. In this study, first a probabilistic bond failure mode prediction model that considers various influencing factors including loading type and corrosion is developed in this study. This study uses the bond testing results of 132 beam-end specimens subjected to monotonic and cyclic loading and adopts classification methods to develop the prediction model, which is then used to evaluate the impact of bond behavior on the reliability of a RC beam with a lap splice. Then, multivariate nonlinear regression with all-possible subset model selection and symbolic multi-gene regression are adopted for probabilistic model development for bond strength and peak slip under the two bond failure modes considering corrosion. In particular, a comprehensive bond dataset collected from bond tests on the beam and beam-end specimens in the literature and from the experimental testing conducted in this study, and a criterion to specify the bond failure (open full item for complete abstract)

Committee: Qindan Huang (Advisor); David Roke (Committee Member); Craig Menzemer (Committee Member); Ping Yi (Committee Member); Richard Einsporn (Committee Member) Subjects: Civil Engineering; Design; Engineering; Experiments

Master of Sciences (Engineering), Case Western Reserve University, 2024, EECS - System and Control Engineering

Wire Arc Additive Manufacturing (WAAM) has emerged as a promising technology for producing metal parts, offering reduced lead times and costs compared to traditional methods. However, achieving optimal process parameters in WAAM and accurately predicting bead height remain challenging due to complex interactions between input variables and output characteristics. This thesis addresses the challenge of developing a machine learning regression model to predict the average bead height of single deposited beads, crucial for building simple and complex shapes in WAAM. The research investigates the relationship between four critical input parameters - Voltage, Wire Feed Speed (WFS), Travel Speed, Contact Tip to Work Distance (CTWD) - and their influence on bead dimensions in WAAM. A comprehensive experimental setup is employed, utilizing a custom-built WAAM 3D metal printer equipped with a gantry system and controlled by a Duet 3 controller. Steel wire ER70s-6 with a diameter of 0.9mm is used for printing, producing single beads with heights ranging from 2.5mm to 3.55mm. A total of 248 experiments are performed using the Arc-One Machine at Case Western Reserve University (CWRU) for the model training, which are then analyzed. A machine learning regression model is built using this dataset, with four inputs (Voltage, Travel Speed, Wire Feed Speed, Contact Tip to Work Distance) and two corresponding outputs (average bead height and variance of bead heights). Various analytical techniques were explored to predict the average bead height and its variance, leading to the adoption of the Gradient 18 Boosting regression model as the most effective approach. Two models, a forward model and an inverse model, were developed to predict WAAM parameters and outputs. The forward model predicts the average bead height and variance based on the input parameters (Voltage, Wire Feed Speed, Travel Speed, and Contact Tip to Work Distance), providing insights into how th (open full item for complete abstract)

Committee: Kenneth Loparo (Committee Chair); Robert Gao (Committee Member); John Lewandowski (Committee Member); Robert Gao (Committee Member); John Lewandowski (Committee Member); Kenneth Loparo (Advisor) Subjects: Aerospace Engineering; Design; Experiments; Materials Science

Doctor of Philosophy (Ph.D.), Bowling Green State University, 2024, Statistics

Deep learning, neural network, has been penetrating into almost every corner of data analyses. With advantages on computing power and speed, adding more layers in a neural network analysis becomes a common practice to improve the prediction accuracy. However, over depleting information in the training dataset may consequently carry data noises into the learning process of neural network and result in over-fitting errors. Neural Network has been used to predict the future time series data. It had been claimed by several authors that the Neural Network (Recurrent Neural Network) can predict the time series data, although time series models have also been used to predict the future time series data. This dissertation is thus motivated to investigate the prediction performances of neural networks versus the conventional inference method of time series analysis. After introducing basic concepts and theoretical background of neural network and time series prediction in Chapter 1, Chapter 2 analyzes fundamental structure of time series, along with estimation, hypothesis testing, and prediction methods. Chapter 3 discusses details of computing algorithms and procedures in neural network with theoretical adjustment for time series prediction. In conjunction with terminologies and methodologies in the previous chapters, Chapter 4 directly compares the prediction results of neural networks and conventional time series for the squared error function. In terms of methodology assessment, the evaluation criterion plays a critical role. The performance of the existing neural network models for time series predictions has been observed. It has been experimentally observed that the time series predictions by time series models are better compared to the neural network models both computationally and theoretically. The conditions for the better performances of the Time Series Models over the Neural Network Models have been discovered. Theorems have also been pro (open full item for complete abstract)

Committee: John Chen Ph.D. (Committee Chair); Hanfeng Chen Ph.D. (Committee Member); Umar Islambekov Ph.D. (Committee Member); Brigid Burke Ph.D. (Other) Subjects: Applied Mathematics; Artificial Intelligence; Behavioral Sciences; Computer Science; Education Finance; Finance; Information Systems; Operations Research; Statistics

PhD, University of Cincinnati, 2020, Arts and Sciences: Mathematical Sciences

Multiple testing on large-scale dependent count data faces challenges in two basic modeling elements, that is, modeling of the dependence structure across observations and the distribution specification on the null and non-null states. We propose three Poisson hidden Markov models (PHMM) under the Bayesian hierarchical model framework to handle these challenges. The dependence structure across hypotheses is modeled through the hidden Markov process. To address the challenge of the distribution specification under the non-null state, several model selection methods are employed and compared to determine the number of mixture components in the non-null distribution. Furthermore, we examine two different ways to include covariate effects, PHMM with homogeneous covariate effects (PHMM-HO) and PHMM with heterogeneous covariate effects (PHMM-HE). Modeling covariate effects helps take consideration of multiple factors which are directly or indirectly related to the hypotheses under investigation. We carry out extensive simulation studies to demonstrate the performance of the proposed hidden Markov models. The stable and robust results show the significant advantages of our proposed models in handling complex data structure in dependent counts. Multiple hypotheses testing with PHMM is valid and optimal compared with a group of commonly used testing procedures. Both PHMM-HO and PHMM-HE improve the multiple testing performance and are able to detect the dynamic data pattern along with the covariate effects.

Committee: Xia Wang Ph.D. (Committee Chair); Hang Joon Kim Ph.D. (Committee Member); Siva Sivaganesan Ph.D. (Committee Member); Seongho Song Ph.D. (Committee Member); Bin Zhang Ph.D. (Committee Member) Subjects: Statistics

Doctor of Philosophy in Urban Studies and Public Affairs, Cleveland State University, 2018, Maxine Goodman Levin College of Urban Affairs

The land bank is a government entity that focuses on the conversion of vacant, abandoned, and tax-delinquent properties into productive use. The object of the land bank is to gain control over the city's problem properties to make possible their timely and productive reuse. The land bank has become a popular policy measure to control the distressed properties in the neighborhood following the foreclosure crisis across in the United States. The objective of this study is to evaluate the spillover effect of the land bank on nearby properties. The primary research question is as follows: has the land bank public intervention created a positive spillover effect on nearby home sales in the respective neighborhood in the City of Cleveland, Ohio? This is a case study for one city. This study utilized the spatial hedonic model to measure the impact of a two-year land bank acquisition period on nearby property values within two buffers: 500 feet and 1,000 feet. This study also utilized the Geographically Weighted Regression to evaluate the local variation of the effect over the space. The study period is 24 months from September 2012 to August 2014. This study identifies that two years of land bank acquisitions have had a positive effect within the 500 feet buffer from the sale location. The pure effect of two years of land bank acquisitions results in a positive 1.82% impact by OLS estimation and a positive 1.81% impact by ML, 2SLS, and 2SLS-robust estimations. The mean value of the implicit marginal price is $897 over 24 months of sale data from September 2012 to August 2014. This estimated benefit may not have existed if the land bank did not acquire the abandoned properties. The result of this study will support policymakers and practitioners in their decision to expand land bank programs.

Committee: Dennis Keating Ph.D. (Committee Chair); William Bowen Ph.D. (Committee Member); Wonseok Seo Ph.D. (Committee Member) Subjects: Area Planning and Development; Public Policy

Doctor of Philosophy, The Ohio State University, 2018, Political Science

A good deal of research in the counterinsurgency literature focuses on the effects of service provision and nation-building programs that incentivize support for the counterinsurgent force. At the same time, many insurgent groups also engage in similar distributions of public goods. The overall consequences of these non-state service programs are much less well understood, and this dissertation seeks to identify these effects by assessing the link between the provision of social services and insurgent success against state forces. When the role of social services is evaluated using traditional statistical methods, it appears that these public goods increase the likelihood that a militant group will perform well against the state. However, the observation of these non-state social service programs is shaped by underlying conflict dynamics that affect the ability to draw conclusions about their role in the outcome of a fight. The groups that choose to provide services and succeed in doing so despite government efforts to stop them are likely to be stronger than those groups that we do not see providing services. The endogeneity of social services therefore calls into question the finding that services make insurgents more successful. As a solution to this challenge, this project applies flexible joint regression modeling, a recently-developed approach for endogenous treatment variables, to assess the impacts of insurgent-provided services in the complex contexts of civil conflict and counterinsurgent operations. Results of this analysis indicate that social service programs implemented by insurgents do not significantly shape the outcome of civil conflict.

Committee: Bear Braumoeller (Advisor); Christopher Gelpi (Committee Member); Jan Pierskalla (Committee Member); Bradley Holland (Committee Member) Subjects: Political Science

Doctor of Philosophy, Case Western Reserve University, 2018, Epidemiology and Biostatistics

Many remarkable advances have been made in the non-parametric and semiparametric methods for high-dimensional longitudinal data. However, there is a lack of a method for addressing missing data in these important methods. Motivated by oxygenation of retinopathy of prematurity (ROP) study, we developed a penalized spline mixed effects model for a high-dimensional nonlinear longitudinal continuous response variable using the Bayesian approach. The ROP study is complicated by the fact that there are non-ignorable missing response values. To address the non-ignorable missing data in the Bayesian penalized spline model, we applied a selection model. Properties of the estimators are studied using Markov Chain Monte Carlo (MCMC) simulation. In the simulation study, data were generated with three different percentages of non-ignorable missing values and three different sample sizes. Parameters were estimated under various scenarios. The proposed new approach did better compare to the semiparametric mixed effects model with non-ignorable missing values under missing at random (MAR) assumption in terms of bias and percent bias in all scenarios of non-ignorable missing longitudinal data. We performed sensitivity analysis for the hyper-prior distribution choices for the variance parameters of spline coefficients on the proposed joint model. The results indicated that half-t distribution with three different degrees of freedom did not influence to the posterior distribution. However, inverse-gamma distribution as a hyper-prior density influenced to the posterior distribution. We applied our novel method to the sample entropy data in ROP study for handling nonlinearity and the non-ignorable missing response variable. We also analyzed the sample entropy data under missing at random.

Committee: Abdus Sattar (Advisor); Mark Schluchter (Committee Chair); Albert Jeffrey (Committee Member); Abdus Sattar (Committee Member); Sana Loue (Committee Member) Subjects: Biostatistics

Master of Arts, Miami University, 2014, Geography

The ability to predict American beech distribution (Fagus grandifolia Ehrh.) from environmental data was tested by using a geographic information system (GIS) in tandem with species distribution models (SDMs). The study was conducted in Butler and Preble counties in Ohio, USA. Topography, soils, and disturbance were approximated through 15 predictor variables with presence/absence and basal area serving as the response variables. Using a generalized linear model (GLM) and a boosted regression tree (BRT) model, curvature, elevation, and tasseled cap greenness were shown to be significant predictors of beech presence. Each of these variables was positively related to beech presence. A linear model using presence only data was not effective in predicting basal area due to a small sample size. This study demonstrates that SDMs can be used successfully to advance our understanding of the relationship between tree species presence and environmental factors. Large sample sizes are needed to successfully model continuous variables.

Committee: Mary Henry PhD (Advisor); David Gorchov PhD (Committee Member); Jerry Green PhD (Committee Member) Subjects: Botany; Ecology; Forestry; Geography

Doctor of Philosophy, The Ohio State University, 2010, City and Regional Planning

Criteria air pollutants, such as nitrogen oxides (NOx), carbon monoxide (CO), sulfur dioxide (SO2), particulate matter (PM), and ozone (O3), are characterized by temporal and locational hot spots in urban areas, frequently violating pollution standards, and, as a result, threatening the health and well-being of the population. Several factors, such as the intensity and duration of emissions, the chemical reactions among pollutants, the uptake and assimilation of pollutants by urban vegetation, and the meteorological factors that induce chemical reactions and atmospheric dispersion, have been considered as explanatory variables in air quality models. Among them, emissions from motor vehicles turn out to be a key determinant of the spatial and temporal patterns of ambient pollution concentrations. The purpose of this research is to formulate and estimate (1) metropolitan-wide time-series air quality models and (2) land-use regression (LUR) air quality panel models, in order to explain spatio-temporal variations in pollution concentrations. Using the Seoul Metropolitan Area as a case study, traffic counts, vehicle-kilometers-traveled (VKT), land uses, and meteorological factors, such as solar radiation, temperature, humidity, wind speed and wind direction, are used as explanatory variables. An extensive understanding of atmospheric pollutants chemistry is reflected in the formulation of these models. Differences in concentrations measured at air quality monitoring stations (AQMs) across the week (weekdays vs. Sunday) and geographical locations (roadside vs. background), are also investigated using dummy variables and the product of these variables with the original variables. The results of the time-series models and panel regression models indicate that traffic counts and VKT are significant in explaining the concentrations of both directly emitted pollutants, such as NO2, CO, SO2, and PM, and O3, a secondary pollutant. The concentrations of the directly emitted poll (open full item for complete abstract)

Committee: Jean-Michel Guldmann PhD (Committee Chair); Steven Gordon PhD (Committee Member); Philip Viton PhD (Committee Member); Gulsah Akar PhD (Committee Member) Subjects: Urban Planning

Doctor of Philosophy (PhD), Ohio University, 2008, Biological Sciences (Arts and Sciences)

This study examined the role of spatial distribution of land cover on water quality and stream fish assemblages in watersheds of low-order streams in the Mississippi coastal plain. I found that the growing proliferation of urbanized land into landscapes with dominant rural or forest character decreases water quality and diversity of aquatic biota. A reconstruction of local land use history identified the contemporary landscape mosaic as legacy of the Southern Lumber Boom and management decisions based on individual land-ownership. Such decisions transformed firedominated longleaf pine savanna into a landscape characterized by active fire suppression and second-growth Southern mixed deciduous forest, non-industrial pine silviculture, and an expanding urban core. Commercial development is concentrated in floodplains and along major transportation routes, while diffuse parcel-size residential development across the study area increases fragmentation of the forest-dominated landscape matrix.Contemporary land cover distribution was evaluated using a new hybrid classification method combining panchromatic aerial photographs, highresolution multispectral remote sensing data, and Landsat5 TM images. A spatially explicit modeling approach using GIS quantified watershed land cover based on distance to streams and relative upstream distance from sampling sites. Water chemistry, stream geomorphology and fish assemblage metrics identified direct and indirect linkages between land cover, landscape features, and stream ecology. In the Mississippi coastal plain land cover influence exceeded geomorphological effects on stream conditions. Fish assemblages varied among sites in composition and diversity, and differed between watersheds with contrasting dominant land cover, suggesting integration of watershed-scale and local-scale influences. Fish assemblage metrics identified species richness, assemblage dominance, trophic guild membership, and perturbation tolerance as best descrip (open full item for complete abstract)

Committee: Glenn R. Matlack PhD (Committee Chair); Kelly Johnson PhD (Committee Member); James Lein PhD (Committee Member); Molly Morris PhD (Committee Member); Philip Cantino PhD (Committee Member) Subjects: Ecology; Geography

Ph.D., Antioch University, 2006, Leadership and Change

The purpose of this study was to identify predictors of leadership using a newly developed assessment for leadership selection within the healthcare industry by comparing assessment scores to supervisor rankings of the subjects. The study population consisted of 195 employees of 11 different hospitals. Each of the participants completed the Healthcare Leadership Inventory (HLI) assessment; their immediate supervisors completed performance ratings for them. None of the instruments were designed by the researcher. The dependent variable of the study was the supervisor-provided factor of Promotion Potential. Stepwise multiple regression was the main analytical approach. The analysis yielded two predictors of leadership success from the HLI assessment (Achievement Orientation and Openness to Change) and five from the Supervisor Ratings (Multi-Tasking, Drive for Results, Self Confidence, Openness to Change, and Customer Orientation). The identified predictors from each instrument had construct symmetry, although they were not statistically duplicative. The predictors from Supervisor Ratings provide some insight into the implicit leadership theories shared by management personnel in the healthcare industry. The HLI assessment factors of Achievement, Conscientiousness, Innovative, and Customer Focus had significant correlations with their counterparts from Supervisor Ratings. The Critical Thinking factor surprisingly did not significantly predict leadership potential or correlate with any of the other factors. The electronic version of the dissertation is accessible at the OhioLINK ETD center http://www.ohiolink.edu/etd/.

Committee: Jon Wergin (Advisor) Subjects:

Doctor of Philosophy, The Ohio State University, 2024, Statistics

Modern computer experiment applications may model a physical system using multiple physics-based simulators. The fidelity of each simulator may vary across the input domain. Rather than selecting a single simulator for inference and prediction, one strategy is to combine the set of simulators under consideration. The most common strategy is to estimate the underlying system by combining, or "mixing", the predictions from K different simulators using a linear combination. The primary objective is to then obtain a better global prediction of the system and gain some insight regarding the fidelity of each simulator. Classical approaches combine the K models using scalar weights. In some cases, these weights may indicate each individual model's overall, or global, predictive performance. More recent work combines the predictions from K models using input-dependent weights. This strategy allows for a more localized prediction and interpretation, as the weight functions may reflect each individual simulator's local fidelity. Local weighting schemes introduce a new challenge in that one must define a relationship between the inputs and the weights. A common choice is to define the weight functions using linear bases, however selecting the appropriate bases is a non-trivial task. This work proposes a Bayesian Additive Regression Tree (BART) model for the weight functions. The weight functions are then defined using tree bases that are adaptively learned based on the information in the data and the model set. Our approach is designed to not only improve global prediction of the system, but also allow for reliable inference regarding where each model is accurate or inaccurate, relative to the others in the model set. Using the additive tree bases, the weight functions are modeled as piecewise constant functions. In some cases, it may be desirable to estimate the weights and resulting mixed prediction as continuous functions. Motivated by this, we propose a ran (open full item for complete abstract)

Committee: Matthew Pratola (Advisor); Christopher Hans (Committee Member); Thomas Santner (Committee Member) Subjects: Statistics

Psy. D., Antioch University, 2023, Antioch Santa Barbara: Clinical Psychology

Hispanic participation in Alzheimer's disease (AD) research studies has been historically low. With low engagement, there are many nuances which are not understood related to AD care in the Hispanic Community. The primary purpose of this study is to analyze a Hispanic data set of risk factors of Alzheimer's Disease. Three predictors have been identified to be highly correlated with the onset of Alzheimer's disease in other populations and will be analyzed to indicate how predictive they are in a diagnosis of Alzheimer's disease in a Hispanic population. This dissertation is available in open access at AURA, https://aura.antioch.edu/ and OhioLINK ETD Center, https://etd.ohiolink.edu

Committee: Brett Kia-Keating Ed.D. (Committee Chair); Stephen Southern Ed.D. (Committee Member); Matthew Nance Ph.D. (Committee Member) Subjects: Health; Hispanic Americans; Psychological Tests; Psychology

MA, University of Cincinnati, 2023, Arts and Sciences: Geography

Drug abuse is an important issue in the United States. Marijuana is one of the most commonly used types of drugs, and the long-time use of marijuana can lead to health problems. But as the medical uses of marijuana have been discussed in recent studies, laws and attitudes toward marijuana in the United States have become more permissive. Some studies have explored the impact of the legalization of medical marijuana on people's attitudes toward drugs and drug-related criminal activities, but gaps remain in research, and the relationship between the legalization of medical marijuana and drug activities is still inconclusive. This study aims to analyze changes in the temporal and spatial distribution of drug activity at the micro research unit and test the impact of the legalization of medical marijuana on it. Three main research questions are explored: (1) Is there a change in the amount of drug activity before and after the legalization of medical marijuana? (2) Is there a change in the concentration of drug activity before and after the legalization of medical marijuana? (3) Is the effect of medical marijuana legalization on drug activity statistically significant, with the socio-economic factors and land-use factors controlled? Ohio became the 25th state in the U.S. to legalize medical marijuana on September 8, 2016. Cincinnati, Ohio is the chosen research area and call for service data collected by the emergency telephone service from 2013 to 2018 is used to count drug activities. Results indicate a significant decrease in drug activities since legalization. The number of street segments with drug activities also decreased, suggesting that drug activities became more concentrated. Medical marijuana legalization has likely had a significant effect on drug activity and it is possible that the legalization of medical marijuana has changed people's attitudes towards marijuana-related activities, which in turn affected the likelihood of reporting via 911 calls. The res (open full item for complete abstract)

Committee: Lin Liu Ph.D. (Committee Chair); Changjoo Kim Ph.D. (Committee Member); Jeffrey Brewer Ph.D. (Committee Member) Subjects: Geographic Information Science

PhD, University of Cincinnati, 2023, Engineering and Applied Science: Computer Science and Engineering

Given a test input, not knowing the expected output of the software under test (SUT) is called the oracle problem. An emerging method of alleviating the oracle problem is metamorphic testing (MT). Rather than focusing on the correctness of output from a single execution of the SUT, MT exploits metamorphic relations (MRs) as derived oracles for checking the functional correctness of the code. Although researchers have argued that MT can be a simple and effective technique to help software developers, little is known about the actual cost of constructing MRs in real-world software and the relationship between MT and the already well-adopted method in software development. This research examines a series of practices to evaluate the effectiveness of MT during software development. Our investigation is conducted within the context of a real-world scientific software, the Storm Water Management Model (SWMM), developed and maintained by the U.S. Environmental Protection Agency (EPA). To ensure SWMM's accuracy in modeling stormwater runoff and executing hydraulic and water quality simulations, the development team continually evolves the software. Among the challenges they face, software testing stands out as one of the most technically complex tasks. Our research initially investigates the current testing practices and software quality assurance (QA) workflows in scientific software development, taking the SWMM as a case study. The value of our work resides in the qualitative characterizations and quantitative assessments of the tests that scientific software developers have independently written and released within the SWMM context. Our findings indicate that oracles indeed play a role in scientific software testing. Furthermore, by employing an empirical approach, we identified four critical requirements for the improved integration of MT into scientific software development. Constructing MRs is critical because without them, MT cannot b (open full item for complete abstract)

Committee: Nan Niu Ph.D. (Committee Chair); Wen-Ben Jone Ph.D. (Committee Member); Tingting Yu Ph.D. (Committee Member); Boyang Wang Ph.D. (Committee Member); Michelle Simon Ph.D. (Committee Member) Subjects: Computer Science

Doctor of Philosophy, University of Toledo, 2023, Mathematics

The number needed to treat (NNT) is an efficacy measure used in randomized clinical trials (RCTs), systematic reviews, and meta-analyses. NNT is defined as the average number of patients that need to be treated to achieve a successful outcome due to treatment. There are mainly three different measures of NNT : Furukawa and Leucht's (2011), Kraemer and Kupfer's (2006), and Laupacis et al.'s (1988). While the first two measures deal with continuous outcomes, the Laupacis method is based on binary outcome variables. Vancak et al. (2020) have proposed three novel NNT estimators to supplement the NNTs introduced by Laupacis, Furukawa and Leucht, and Kraemer and Kupfer. In this study, we aim to estimate different measures of NNT under a semiparametric density ratio model. First, by assuming a two-sample density ratio model of the treatment and control populations, we propose a semiparametric distribution function estimator. We derive the asymptotic distributions of the new semiparametric distribution function estimators and demonstrate its higher efficiency compared to its nonparametric counterparts. Furthermore, we present the joint distribution of the semiparametric distribution function estimators and compare its efficiency to the joint distribution of the nonparametric estimators. Second, we propose a semiparametric estimator for Laupacis' NNT under a semiparametric density ratio model. We derive the large sample properties of our proposed estimator and demonstrate its higher efficiency compared to its nonparametric counterpart. The proposed semiparametric estimator is shown, via a simulation study, to be more robust than a fully parametric approach and is more accurate than a fully nonparametric approach. In addition, we present some results from the analysis of two real examples. Finally, when the outcome is continuous, we propose a semiparametric approach to estimate Kraemer and Kupfer's NNT. Under a semiparametric density ratio model, (open full item for complete abstract)

Committee: Biao Zhang (Committee Chair); Geoffrey Martin (Committee Member); Tian Chen (Committee Member); Rong Liu (Committee Member) Subjects: Statistics

Doctor of Philosophy (Ph.D.), Bowling Green State University, 2023, Statistics

Since the advent of high-dimensional data structures in many areas such as medical and biological sciences, economics, and marketing investigation over the past few decades, the need for statistical modeling techniques of such data has grown. In high-dimensional statistical modeling techniques, model selection is an important aspect. The purpose of model selection is to select the most appropriate model from all possible high-dimensional statistical models where the number of explanatory variables is larger than the sample size. In high-dimensional model selection, endogeneity is a challenging issue. Endogeneity is defined as when a predictor variable (X) in a regression model is related to the model error term (ϵ), which results in inconsistency of model selection. Because of the existence of endogeneity, Fan and Liao (2014) pointed out that exogenous assumptions in most statistical methods are not able to validate in high-dimensional model selection, and exogenous assumptions means a predictor variable (X) in a regression model is not related to the model error term (ϵ). To avoid the effect of endogeneity, Fan and Liao (2014) proposed the focused generalized method-of-moments (FGMM) approach in high-dimensional linear models with endogeneity for selecting significant variables consistently. We propose the FGMM approach with modifications for high-dimensional linear and nonlinear models with endogeneity to choose all of the significant variables. The theorems in Fan and Liao (2014) show that FGMM approach consistently chooses the true model as the sample size goes to infinity in both the linear and nonlinear models. In linear models with endogeneity, we modify the penalty term to improve the selection performance. In nonlinear models with endogeneity, we adjust the loss function in the FGMM approach to achieve model selection consistency, which is to select the true model as the sample size n goes to infinity. This modified approach adopts inst (open full item for complete abstract)

Committee: Junfeng Shang Ph.D (Committee Chair); Meagan Docherty Ph.D (Other); John Chen Ph.D (Committee Member); Wei Ning Ph.D (Committee Member) Subjects: Statistics

Master of Science (MS), Bowling Green State University, 2022, Applied Statistics (Math)

One of the most important features of collected data is the identification of change points. Researchers are often tasked with identifying abrupt or any changes that appeared in a dataset. Identifying these changes and extracting meaningful information is very important to benefit and avoid losses. In the early 1970's Hirotugu Akaike formulated what is now called the Akaike Information Criterion (AIC). After that, in 1978, Schwarz introduced the improved version of the AIC called Schwarz Information Criterion (SIC) which served as an asymptotic approximation to a transformation of the Bayesian posterior probability of a candidate model. In this thesis, statistical analysis of regression models related to the application of modified information criterion (MIC) introduced by Chen et al (2006) will be explored along with the comparison of SIC. Chapters 2 and 3 consists of the derivations of theoretical foundation and mathematical formulae. The regression model is employed in each situation. Simulations have been conducted for each information criterion to compare the performance of the two approaches. After the comparison of the performances, the application of the MIC has been employed on three different datasets out of which two have already been tested for finding the change point locations. The only difference between these two information criteria is in their penalty terms. The objective of the thesis is to compare the performance of these two approaches and use the better one to extract meaningful information after the identification of change point locations.

Committee: Wei Ning Ph.D. (Committee Chair); John Chen Ph.D. (Committee Member) Subjects: Mathematics; Statistics

MS, University of Cincinnati, 2021, Medicine: Biostatistics (Environmental Health)

Bone damping is a non-invasive measure of bone fragility and is identified as a better predictor of osteoporosis (OP) related bone fracture/fragility than bone mineral density (BMD). Subject with higher damping value demonstrates a heightened resistance to fracture. The purpose of this study was to identify the predictors of bone shock absorption (BSA) capacity measured as a damping factor by using model selection and multivariate multiple regression (MMR) method as well as regression tree. The main dataset was from an existing Cincinnati Lead Study (CLS) cohort. It is a prospective and longitudinal study examined early and late effects of childhood lead exposure on growth and development. The results of this study indicated that cortical vBMD, cortical thickness, endosteal circumference, cortical section modulus, current weight, and the number of pregnancies carried until the 3rd trimester were significant predictors of bone damping factor based on the method of model selection and MMR. Among the predictors with top nine highest variable importance values in regression tree method, four are the same as significant predictors from MMR analysis. Those are current weight, cortical section modulus, cortical vBMD, and endosteal circumference. Cortical section modulus and cortical vBMD have positive relationship with damping factor; however weight and endosteal circumference have negative relationship with damping factor. All variables' relationships with damping factor are clinically significant. Lack of dataset from normal people to compare the differences and the missingness of the data are the limitation of the study. Current weight, cortical section modulus, cortical vBMD, and endosteal circumference are significant predictor of damping factor based on the study results. They are biologically relevant to damping and statistically significant in the damping model.

Committee: Amit Bhattacharya (Committee Member); Marepalli Rao Ph.D. (Committee Chair) Subjects: Biostatistics