Statistics and Probability Commons^™

Analyzing Competitive Balance In Professional Sport, Kevin Alwell

Honors Scholar Theses

In this paper we review several measures to statistically analyze competitive balance and report which leagues have a wider variance of performance amongst its competitors. Each league seeks to maintain high levels of parity, making matches and overall season more unpredictable and appealing to the general audience. Here we quantify competitive advantage across major sports leagues in numbers using several statistical methods in order for leagues to optimize their revenue.

Using Saddlepoint Approximations And Likelihood-Based Methods To Conduct Statistical Inference For The Mean Of The Beta Distribution, Bryn Brakefield

Electronic Theses and Dissertations

The prevalence of conducting statistical inference for the mean of the beta distribution has been rising in various fields of academic research, such as in immunology that analyzes proportions of rare cell population subsets. For our purposes, we will address this statistical inference problem by using likelihood-based applications to hypothesis testing, along with a relatively new statistical method called saddlepoint approximations. Through simulation work, we will compare the performance of these statistical procedures and provide both the statistical and scientific communities with recommendations on best practices.

Predicting The Federal Funds Rate, Danielle Herzberg

Undergraduate Theses and Capstone Projects

This thesis examines various economic indicators to select those that are the most significant in a predictive model of the Effective Federal Funds Rate. Three different statistical models were built to show how monetary policy changed over time. These three models frame the last economic downturns in the United States; the tech bubble, the housing bubble, and the Great Recession. Many iterations of statistical regressions were conducted in order to achieve the final three models that highlight variables with the highest levels of significance. It is important to note the economic data has high levels of autocorrelation, and that these …

Life And Death: Quantifying The Risk Of Heart Disease With Machine Learning, Jack Scott Glienke

Honors Program Theses

Coronary heart disease has long been a key area of focus in the discussion of public health. As such, numerous studies have been conducted throughout history with the sole intention of identifying risk factors leading to the onset of cardiovascular conditions. A plethora of statistical procedures can be used to identify an individual’s risk of developing heart disease, yet regression models tend to be the default tool used by researchers. Using the data obtained from the most influential cardiovascular study to date, the Framingham Heart Study, this analysis uses machine learning techniques to generate and test the predictive power of …

Predicting Disease Progression Using Deep Recurrent Neural Networks And Longitudinal Electronic Health Record Data, Seunghwan Kim

McKelvey School of Engineering Graduate Student Theses & Dissertations

Electronic Health Records (EHR) are widely adopted and used throughout healthcare systems and are able to collect and store longitudinal information data that can be used to describe patient phenotypes. From the underlying data structures used in the EHR, discrete data can be extracted and analyzed to improve patient care and outcomes via tasks such as risk stratification and prospective disease management. Temporality in EHR is innately present given the nature of these data, however, and traditional classification models are limited in this context by the cross- sectional nature of training and prediction processes. Finding temporal patterns in EHR is …

Applications Of Machine Learning In High-Frequency Trade Direction Classification, Jared E. Hansen

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

The correct assignment of trades as buyer-initiated or seller-initiated is paramount in many quantitative finance studies. Simple decision rule methods have been used for signing trades since many data sets available to researchers do not include the sign of each trade executed. By utilizing these decision rule methods, as well as engineering new variables from available data, we have demonstrated that machine learning models outperform prior methods for accurately signing trades as buys and sells, achieving state-of-the-art results. The best model developed was 4.5 percentage points more accurate than older methods when predicting onto unseen data. Since finance and economics …

Motivational Predictors Of Academic Risk-Taking, Danette Dee Barber

UNLV Theses, Dissertations, Professional Papers, and Capstones

Students benefit when they are willing to engage in optimal challenges (Clifford, 1991). Engagement in challenges, however, comes with academic risks, as failure may be a result. This study investigated motivational factors, including expectancy, subjective task value, mastery goal orientation, and performance avoidance goal orientation as predictors of achievement-related outcomes, including course grade and academic risk-taking. Data were collected from 317 university students enrolled in education classes. Students were given a reading passage and asked to choose questions to answer based on the passage. Students who chose harder questions were categorized as taking more risk. Students also answered questions about …

Novel Bayesian Methodology For The Analysis Of Single-Cell Rna Sequencing Data., Michael Sekula

Electronic Theses and Dissertations

With single-cell RNA sequencing (scRNA-seq) technology, researchers are able to gain a better understanding of health and disease through the analysis of gene expression data at the cellular-level; however, scRNA-seq data tend to have high proportions of zero values, increased cell-to-cell variability, and overdispersion due to abnormally large expression counts, which create new statistical problems that need to be addressed. This dissertation includes three research projects that propose Bayesian methodology suitable for scRNA-seq analysis. In the first project, a hurdle model for identifying differentially expressed genes across cell types in scRNA-seq data is presented. This model incorporates a correlated random …

Introduction To Research Statistical Analysis: An Overview Of The Basics, Christian Vandever

HCA Healthcare Journal of Medicine

This article covers many statistical ideas essential to research statistical analysis. Sample size is explained through the concepts of statistical significance level and power. Variable types and definitions are included to clarify necessities for how the analysis will be interpreted. Categorical and quantitative variable types are defined, as well as response and predictor variables. Statistical tests described include t-tests, ANOVA and chi-square tests. Multiple regression is also explored for both logistic and linear regression. Finally, the most common statistics produced by these methods are explored.

Rdc Data Alternatives: Conducting Research During Covid-19, Kristi Thompson, Elizabeth Hill

Western Libraries Presentations

Recent physical distancing protocols pertaining to the COVID-19 Pandemic have meant that RDC researchers need to find alternatives ways of carrying out their research. The Real Time Remote Access (RTRA) program offers one alternative way to access confidential Statistics Canada data. Other options include using the Statistics Canada public use files and analyzing data from other sources.

The presenters, data librarians from Western Libraries will discuss the differences between the data that can be accessed through the RTRA the RDC. RTRA data is a very useful option for some types of questions but also has some important limitations. We will …

Accurate Confidence Intervals For Risk Difference In Meta-Analysis With Rare Events, Tao Jiang, Baixin Cao, Guogen Shan

Environmental & Global Health Faculty Research

Background: Meta-analysis provides a useful statistical tool to effectively estimate treatment effect from multiple studies. When the outcome is binary and it is rare (e.g., safety data in clinical trials), the traditionally used methods may have unsatisfactory performance. Methods: We propose using importance sampling to compute confidence intervals for risk difference in meta-analysis with rare events. The proposed intervals are not exact, but they often have the coverage probabilities close to the nominal level. We compare the proposed accurate intervals with the existing intervals from the fixed- or random-effects models and the interval by Tian et al. (2009). Results: We …

Effect Of Zinc On Microcystis Aeruginosa And Its Toxin Production, Jose L. Perez

Seton Hall University Dissertations and Theses (ETDs)

Cyanobacteria harmful algal blooms (CHABs) are globally increasing biomasses of detrimental cyanobacteria due to anthropogenic water phosphorous and nitrogen loading and climate change. CHABs often produce secondary metabolites, cyanotoxins, that cause harmful effects to organisms, most water systems, and socioeconomic infrastructures. Additionally, the presence of heavy metal pollutant runoff in CHAB affected environments may result in CHAB population changes – aggravating toxigenicity. Zinc metal resistance and stress response were studied in microcystin (MC) cyanotoxin-producing Microcystis aeruginosa UTEX LB 2385 (M. aeruginosa UTEX LB 2385) and non-MC producing Microcystis aeruginosa UTEX LB 2386 (M. aeruginosa UTEX LB 2386) cyanobacteria. Molecular analysis …

Logistic Growth Modeling With Markov Chain Monte Carlo Estimation, Jaehwa Choi, Jinsong Chen, Jeffrey R. Harring

Journal of Modern Applied Statistical Methods

A new growth modeling approach is proposed to can fit inherently nonlinear (i.e., logistic) function without constraint nor reparameterization. A simulation study is employed to investigate the feasibility and performance of a Markov chain Monte Carlo method within Bayesian estimation framework to estimate a fully random version of a logistic growth curve model under manipulated conditions such as the number and timing of measurement occasions and sample sizes.

A Simulation Study On Increasing Capture Periods In Bayesian Closed Population Capture-Recapture Models With Heterogeneity, Ross M. Gosky, Joel Sanqui

Journal of Modern Applied Statistical Methods

Capture-Recapture models are useful in estimating unknown population sizes. A common modeling challenge for closed population models involves modeling unequal animal catchability in each capture period, referred to as animal heterogeneity. Inference about population size N is dependent on the assumed distribution of animal capture probabilities in the population, and that different models can fit a data set equally well but provide contradictory inferences about N. Three common Bayesian Capture-Recapture heterogeneity models are studied with simulated data to study the prevalence of contradictory inferences is in different population sizes with relatively low capture probabilities, specifically at different numbers of …

Doubling Time Of The Covid-19 Epidemic By Province, China, Kamalich Muniz-Rodriguez, Gerardo Chowell, Chi-Hin Cheung, Dongyu Jia, Po-Ying Lai, Yiseul Lee, Manyun Liu, Sylvia Ofori, Kimberlyn M. Roosa, Lone Simonsen, Cecile Viboud, Isaac Fung

Biostatistics, Epidemiology & Environmental Health Sciences: Faculty Publications

In China, the doubling time of the coronavirus disease epidemic by province increased during January 20–February 9, 2020. Doubling time estimates ranged from 1.4 (95% CI 1.2–2.0) days for Hunan Province to 3.1 (95% CI 2.1–4.8) days for Xinjiang Province. The estimate for Hubei Province was 2.5 (95% CI 2.4–2.6) days.

Forecasting San Francisco Bay Area Rapid Transit (Bart) Ridership, Swee K. Chew, Alec Lepe, Aaron Tomkins, Peter Scheirer

SMU Data Science Review

In this paper, we present a forecasting analysis of the San Francisco Bay Area Rapid Transit (BART) ridership data utilizing a number of different time series methods. BART is a major public transportation system in the Bay Area and it relies heavily on its riders' fares; having models that generate accurate ridership numbers better enables the agency to project revenue and help manage future expenses. For our time series modeling, we utilized autoregressive integrated moving average (ARIMA), deep neural networks (DNN), state space models, and long short-term memory (LSTM) to predict monthly ridership. As there is such a wide range …

483— Effectiveness Of Mmr Vaccination In Orthodox Jewish Neighborhoods, Meenu Mundackal

GREAT Day Posters

Measles is a highly contagious disease, where large outbreaks arise by direct contact between susceptible (unvaccinated) and infectious individuals. Many Orthodox Jewish neighborhoods were affected by measles from 2018-2019. To quantify the vaccination effort on this susceptible population, a retrospective analysis was used to study the NYC and Rockland County populations using a differential equations model. A subsequent model, known as a realistically-structured network model, studied only the NYC population, in relation to typical household size. Vaccination strategies were applied to three cohorts: unvaccinated family members, members with 1 prior MMR dose, and members with 2 prior MMR doses. The …

484— Modeling Social Distancing Methods And Their Effectiveness In Combating The Spread Of Ebola, Rachel Fair

GREAT Day Posters

Ebola Virus Disease (EVD) is a rare but severe disease that is transmitted among humans through direct-contact with, and close proximity to, infected bodily fluids. From 2014-16, West Africa experienced the largest Ebola outbreak ever recorded, infecting over 28,000 people, and killing over 11,000. Although the symptoms of EVD are treatable, the disease can be extremely deadly, with an average of 50% EVD cases resulting in fatality. In areas where healthcare is scarce and vaccinations are not readily available, the practices of social distancing and self-quarantining have been shown to be highly effective in combating the spread of EVD. To …

465— Modeling Vaccine Efficacy For Tuberculosis In A Prison Population, Kaitlyn Mundackal

GREAT Day Posters

Tuberculosis is a highly contagious disease and is particularly problematic in confined communities such as prisons. I simulated how Tuberculosis moves through a prison population and tested how much vaccination effort is needed to control its spread. To explore this, I tested adding ever increasing numbers of randomly placed edges in a network and determined the size of the largest component. Afterwards, I removed edges in the model using two different methods, one illustrating if the edges were removed randomly and the other starting with prisoners that had the most connections, to simulate the effect of vaccination. My results show …

Universal Vector Neural Machine Translation With Effective Attention, Joshua Yi, Satish Mylapore, Ryan Paul, Robert Slater

SMU Data Science Review

Neural Machine Translation (NMT) leverages one or more trained neural networks for the translation of phrases. Sutskever intro- duced a sequence to sequence based encoder decoder model which be- came the standard for NMT based systems. Attention mechanisms were later introduced to address the issues with the translation of long sen- tences and improving overall accuracy. In this paper, we propose two improvements to the encoder decoder based NMT approach. Most trans- lation models are trained as one model for one translation. We introduce a neutral/universal model representation that can be used to predict more than one language depending on …