Statistics and Probability Commons^™

Automating Construction And Selection Of A Neural Network Using Stochastic Optimization, Jason Lee Hurt

UNLV Theses, Dissertations, Professional Papers, and Capstones

An artificial neural network can be used to solve various statistical problems by approximating a function that provides a mapping from input to output data. No universal method exists for architecting an optimal neural network. Training one with a low error rate is often a manual process requiring the programmer to have specialized knowledge of the domain for the problem at hand.

A distributed architecture is proposed and implemented for generating a neural network capable of solving a particular problem without specialized knowledge of the problem domain. The only knowledge the application needs is a training set that the network …

Statistical Analysis Of Fatalities Due To Vehicle Accidents In Las Vegas, Nv, Annabelle Marie Mathis

UNLV Theses, Dissertations, Professional Papers, and Capstones

The goal of this thesis is to investigate factors that affect the odds of having a fatality in a vehicle collision. We will be looking at characteristics of the driver that caused the accident (age, gender, behavior, actions, influences, and seat belt worn), the characteristics of the vehicle the driver drove (type of vehicle, and air bag deployment), the characteristics of the environment in which the accident occurred (weather, road condition, lighting, time of day, the day of the week, and month of the year), the characteristics of the crash (direction of accident and how many vehicles were involved), and …

A Comparison Of Spatio-Temporal Prediction Methods Of Cancer Incidence In The U.S, Michelle Hamlyn

UNLV Theses, Dissertations, Professional Papers, and Capstones

Cancer is the cause of one out of four deaths in the United States, and in 2009, researchers expected over 1.5 million new patients to be diagnosed with some form of cancer. People diagnosed with cancer, whether a common or rare type, need to undergo treatments, the amount and kind of which will depend on the severity of the cancer. So how do healthcare providers know how much funding is needed for treatment? What would better enable a pharmaceutical company to determine how much to allocate for research and development of drugs, the amount of each drug to manufacture, or …

Management And Support Of Shared Integrated Library Systems, Jason Vaughan, Kristen Costello

Library Faculty Research

The University of Nevada, Las Vegas (UNLV) University Libraries has hosted and managed a shared integrated library system (ILS) since 1989. The system and the number of partner libraries sharing the system has grown significantly over the past two decades. Spurred by the level of involvement and support contributed by the host institution, the authors administered a comprehensive survey to current Innovative Interfaces libraries. Research findings are combined with a description of UNLV’s local practices to provide substantial insights into shared funding, support, and management activities associated with shared systems.

Statistical Inference Of A Measure For Two Binomial Variates, Serena Petersen

UNLV Theses, Dissertations, Professional Papers, and Capstones

We study measures of a comparison for two independent binomial variates which frequently occur in real situations. An estimator for measure of reduction (MOR) is considered for two sample proportions based on a modified maximum likelihood estimation. We study the desirable properties of the estimator: the asymptotic behavior of its unbiasedness and the variance of the estimator. Since the measure ρ is approximately normally distributed when sample sizes are sufficiently large, one may establish approximate confidence intervals for the true value of the estimators. For numerical study, the Monte Carlo experiment is carried out for the various scenarios of two …

Implementation Of Numerically Stable Hidden Markov Model, Usha Ramya Tatavarty

UNLV Theses, Dissertations, Professional Papers, and Capstones

A Hidden Markov model (HMM) is a statistical Markov model in which the system being modeled is assumed to be a Markov process with unobserved (hidden) states. HMM is an extremely flexible tool and has been successfully applied to a wide variety of stochastic modeling tasks. One of the first applications of HMM is speech recognition. Later they came to be known for their applicability in handwriting recognition, part-of-speech tagging and bio-informatics.

In this thesis, we will explain the mathematics involved in HMMs and how to efficiently perform HMM computations using dynamic programming (DP) which makes it easy to implement …

Calcium Supplemental Usage And Potential Health Issues Sssociated With The Rate Of Usage In Las Vegas, Nv, Tanesha Nicole Moss

UNLV Theses, Dissertations, Professional Papers, and Capstones

Calcium is a very important nutrient and as such, it is very important that all humans consume sufficient amounts. However, some calcium supplements have been known to contain small quantities of lead. This research project used a retrospective approach to explore the trade-off between the benefits of calcium vs. the potential lead exposure amongst people who are taking these supplements. A survey consisting of 10 questions was used to try to assess the rate of consumption of specific types of calcium supplements. This research project obtained lead levels in calcium supplements from previous research and applied that data into this …

Arima Models For Bank Failures: Prediction And Comparison, Fangjin Cui

UNLV Theses, Dissertations, Professional Papers, and Capstones

The number of bank failures has increased dramatically over the last twenty-two years. A common notion in economics is that some banks can become "too big to fail." Is this still a true statement? What is the relationship, if any, between bank sizes and bank failures? In this thesis, the proposed modeling techniques are applied to real bank failure data from the FDIC. In particular, quarterly data from 1989:Q1 to 2010:Q4 are used in the data analysis, which includes three major parts: 1) pairwise bank failure rate comparisons using the conditional test (Przyborowski and Wilenski, 1940); 2) development of the …

Modeling Mortality Rates For Leukemia Between Men And Women In The United States, Blessed Quansah

UNLV Theses, Dissertations, Professional Papers, and Capstones

Leukemia related deaths increased dramatically over the last forty years. Leukemia is a malignant disease or cancer of the bone marrow and blood. It is characterized by the uncontrolled accumulation of blood cells. Leukemia is divided into two categories: myelogenous or lymphocytic, each of which can be acute or chronic. The terms, myelogenous or lymphocytic denote the cell type involved.

In this thesis, the proposed modeling techniques are applied to leukemia deaths data from the Surveillance Epidemiology and End Results (SEER). In particular, annual deaths data from 1969 to 2007 are used in the data analysis, which includes three major …

Effects Of Inlet Conditions On Diffuser Outlet Performance, Zaccary A. Poots

UNLV Theses, Dissertations, Professional Papers, and Capstones

Building air distribution terminal system designers and system installers require accurate quantitative information on the performance of the installed system to achieve optimum efficiency and levels of human comfort. This requires field installation adjustment values from published ideal pressure loss, air distribution and sound generation installation performance. This study documents the air output performance of different installation configurations of six types of ceiling diffusers and compares the results to performance when installed according to ANSI/ASHRAE Standard 70-2006. A diffuser inlet supply plenum was designed for optimum flow and used to acquire a baseline set of data covering the six types …

A Statistical Model For Long-Term Forecasting Of Strong Sand Dust Storms, Siqi Tan

UNLV Theses, Dissertations, Professional Papers, and Capstones

Dust elevated into the atmosphere by dust storms has numerous environmental consequences. These include contributing to climate change; modifying local weather conditions; producing chemical and biological changes in the oceans; and affecting soil formation, surface water, groundwater quality, crop growth, and survival (Goudie and Middleton, 1992). Societal impacts include disruptions to air, road and rail traffic; interruption of radio services; the myriad effects of static-electricity generation; property damage; and health effects on humans and animals (Warner, 2004).

In this thesis, we extend the idea of empirical recurrence rate (ERR), developed by Ho (2008), to model the temporal trend of the …

Risk Auto Theft: Predicting Spatial Distributions Of Crime Events, Tana J. Gurule, Tamara D. Madensen

Graduate Research Symposium (2010 - 2017)

Police typically rely on retrospective hotspot maps to informe prevention strategies aimed at reducing future crime. The current study reviews environmental crime theories that help to identify casual factors associated with rish of auto theft. Map layers are created from data that operationalize these risk factors. These layers are combined using spatial analysis techniques to produce a "risk density" map. Analysis of crime data are used to determing wheter our "risk density" map better predicts subsequetnt theft events than a traditional retrospective hotspot map.

Analysis Of Morris Water Maze Data With Bayesian Statistical Methods, Maxym V. Myroshnychenko, Anton Westveld, Jefferson Kinney

Festival of Communities: UG Symposium (Posters)

Neuroscientists commonly use a Morris Water Maze to assess learning in rodents. In his kind of a maze, the subjects learn to swim toward a platform hidden in opaque water as they orient themselves according to the cues on the walls. This protocol presents a challenge to statistical analysis, because an artificial cut-off must be set for those experimental subjects that do not reach the platform so as they do not drown from exhaustion. This fact leads to the data being right censored. In our experimental data, which compares learning in rodents that have chemically induced symptoms of schizophrenia to …

Relationship Between Perceived And Actual Quality Of Data Checking, Hunter Speich, Sophia Karas, Dan Erosa, Kelly Grob, Kimberly A. Barchard

Festival of Communities: UG Symposium (Posters)

Data quality is critical to reaching correct research conclusions. Researchers attempt to ensure that they have accurate data by checking the data after it has been entered. Previous research has demonstrated that some methods of data checking are better than others, but not all researchers use the best methods. Perhaps researchers continue to use less optimal data checking methods because they mistakenly believe that they are highly accurate. The purpose of this study was to examine the relationship between perceived data quality and actual data quality. A total of 29 participants completed this study. Participants checked that letters and numbers …

Unlv Enrollment Forecasting, Sabrina Beckman, Stefan Cline, Monika Neda

Festival of Communities: UG Symposium (Posters)

Our project investigates the future enrollment of undergraduates at UNLV in the entire university, the College of Science, and the Department of Mathematical Sciences. The method used for the forecast, is the well-known least-squares method, for which a mathematical description will be presented. Studies for the numerical error are pursued too. The study will include graphs that describe the past and future behavior for different parameter settings. Mathematical results obtained show that the university will continue to grow given the current trends of enrollment.

Developing A Library Value Indicator For A Disciplinary Population, Jeanne M. Brown

Library Faculty Research

Three different ways of documenting library value were presented to fourth year landscape architecture students in the UNLV School of Architecture: a contingent valuation survey, a library calculator, and a survey to rate importance and impact of library services and features. Students used the three approaches, then discussed their experiences with the author. Their input suggested improvements in the instruments and provided feedback on possible positive and negative consequences of inviting this kind of valuing. Working with a focused collection and population provided a relatively safe environment to explore concerns about negative consequences.

Demonstrating Library Value: Examples And Applications For Arts Libraries, Jeanne M. Brown

Library Faculty Research

Demonstrating library value is of critical importance to all libraries, both to protect services and to serve patrons effectively. This paper presents suggestions for art and architecture libraries as they engage in determining what patrons value and documenting that value for library and campus administrators. Methods for calculating worth and for presenting a case are provided, as are ways of using strategic thinking and the assessment process to ensure the continuance of valuable services should budget reductions be unavoidable.

Poisson Process Monitoring, Test And Comparison, Qing Chen

UNLV Theses, Dissertations, Professional Papers, and Capstones

The task of determining whether a sudden change occurred in the generative parameters of a time series generates application in many areas. In this thesis, we aim at monitoring the change-point of a Poisson process by method, which is characterized by a forward-backward testing algorithm and several overall error control mechanisms. With the application of this proposed method, we declare that Mount Etna is not a simple Poissonian volcano, because two different regimes divided by the change point, January 30th 1974, are identified. The validation procedures, used in a complementary fashion, by the formal hypothesis tests and graphical method will …

The Determinants Of Colorectal Cancer Survival Disparities In Nevada, Lucas N. Wassira

UNLV Theses, Dissertations, Professional Papers, and Capstones

Different population groups across Nevada and throughout the United States suffer disproportionately from colorectal cancer and its after-effects. Overcoming cancer health disparities is important for lessening the burden of cancer. There has been an overall decline in the incidence of and mortality from colorectal cancer (CRC). This is likely due, in part, to the increasing use of screening procedures such as Fecal Occult Blood Test (FOBT) and/or endoscopy, which can reduce the risk of CRC mortality by fifty percent. Nevertheless, screening procedures are routinely used by only fifty percent of Americans aged fifty years and older. Despite overall mortality decreasing …

Developing A Library Value Indicator For A Disciplinary Population, Jeanne M. Brown

Library Faculty Presentations

Population
- Landscape architecture studio of ten 5th year students
- Use of physical library ranges from 1- 30 times/month
- Use of virtual library ranges from 2-30x/month
- Compared to others in School of Architecture use is moderate
- They self-rate as average or above average on library skills, compared to their peers

Arima Model For Forecasting Poisson Data: Application To Long-Term Earthquake Predictions, Wangdong Fu

UNLV Theses, Dissertations, Professional Papers, and Capstones

Earthquakes that occurred worldwide during the period of 1896 to 2009 with magnitude greater than or equal to 8.0 on the Richter scale are assumed to follow a Poisson process. Autoregressive Integrated Moving Average models are presented to fit the empirical recurrence rates, and to predict future large earthquakes. We show valuable modeling and computational techniques for the point processes and time series data. Specifically, for the proposed methodology, we address the following areas: data management and graphic presentation, model fitting and selection, model validation, model and data sensitivity analysis, and forecasting.

General Coupon Collecting Models And Multinomial Games, James Y. Lee

UNLV Theses, Dissertations, Professional Papers, and Capstones

The coupon collection problem is one of the most studied problems in statistics. It is the problem of collecting r (r<∞) distinct coupons one by one from k different kinds (k<∞) of coupons. We note that this is equivalent to the classical occupancy problem which involves the random allocation of r distinct balls into k distinct cells. Although the problem was first introduced centuries ago, it is still actively investigated today. Perhaps its greatest feature is its versatility, numerous approaches, and countless variations. For this reason, we are particularly interested in creating a classification system for the many generalizations of the coupon collection problem. In this thesis, we will introduce models that will be able to categorize these generalizations. In addition, we calculate the waiting time for the models under consideration. Our approach is to use the Dirichlet Type II integral. We compare our calculations to the ones obtained through Monte Carlo simulation. Our results will show that our models and the method used to find the waiting times are ideal for solving problems of this type.

Insights Into The Commons On Flickr, Jason Vaughan

Library Faculty Research

The Commons on Flickr, comprised of an international community of select libraries, museums, and archives, was a project initially launched in 2008 by the Library of Congress and Flickr. Primary goals of The Commons are to broaden exposure to rich cultural heritage photographs and to observe and participate in the communities of engagement and dialog enabled through The Commons. A survey was administered to all The Commons institutions during summer 2009, focusing on assessment of the overall satisfaction of current members and seeking additional details on participation goals, social interactions, staff time involvement, and general statistics. Members report a very …

Research Poster: Climate Prediction Downscaling Of Temperature And Precipitation In The Great Basin Region, Ramesh Vellore, Benjamin J. Hatchett, Darko Koracin

2010 Annual Nevada NSF EPSCoR Climate Change Conference

Research poster

Research Poster: Hydrological Impacts Of Climate Change On Colorado Basin, Peng Jiang, Zhongbo Yu

2010 Annual Nevada NSF EPSCoR Climate Change Conference

Research poster

Research Poster: An Overview Of Progress In Nsf Epscor Project Entitled, “Reducing Cloud Uncertainties In Climate Models”, Subhashree Mishra, David L. Mitchell, W. Patrick Arnott

2010 Annual Nevada NSF EPSCoR Climate Change Conference

Research poster

In Step With Hiv Vaccines? A Content Analysis Of Local Recruitment Campaigns For An International Hiv Vaccine Study, Paula M. Frew, Wendy Macias, Kayshin Chan, Ashley Harding

Environmental & Global Health Faculty Research

During the past two decades of the HIV/AIDS pandemic, several recruitment campaigns were designed to generate community involvement in preventive HIV vaccine clinical trials. These efforts utilized a blend of advertising and marketing strategies mixed with public relations and community education approaches to attract potential study participants to clinical trials (integrated marketing communications). Although more than 30,000 persons worldwide have participated in preventive HIV vaccine studies, no systematic analysis of recruitment campaigns exists. This content analysis study was conducted to examine several United States and Canadian recruitment campaigns for one of the largest-scale HIV vaccine trials to date (the “Step …

Statistical Inferences For Functions Of Parameters Of Several Pareto And Exponential Populations With Application In Data Traffic, Sumith Gunasekera

UNLV Theses, Dissertations, Professional Papers, and Capstones

In this dissertation, we discuss the usability and applicability of three statistical inferential frameworks--namely, the Classical Method, which is sometimes referred to as the Conventional or the Frequentist Method, based on the approximate large sample approach, the Generalized Variable Method based on the exact generalized p -value approach, and the Bayesian Method based on prior densities--for solving existing problems in the area of parametric estimation. These inference procedures are discussed through Pareto and exponential distributions that are widely used to model positive random variables relevant to social, scientific, actuarial, insurance, finance, investments, banking, and many other types of observable phenomena. …

Chronic Disease, Homeland Security, And Sailing To Where There Be Dragons, David M. Hassenzahl

Public Policy and Leadership Faculty Research

The five papers in this special issue share the perspective that attitudes toward risk are strongly shaped by social context, and that understanding context can help us understand how risk decisions are made, and thereby how to make them better.

Implementation Of Uncertainty Propagation In Triton/Keno, Charlotta Sanders, Denis Beller

Reactor Campaign (TRP)

Monte Carlo methods are beginning to be used for three dimensional fuel depletion analyses to compute various quantities of interest, including isotopic compositions of used nuclear fuel. The TRITON control module, available in the SCALE 5.1 code system, can perform three-dimensional (3-D) depletion calculations using either the KENO V.a or KENO-VI Monte Carlo transport codes, as well as the two-dimensional (2-D) NEWT discrete ordinates code. To overcome problems such as spatially nonuniform neutron flux and non-uniform statistical uncertainties in computed reaction rates and to improve the fidelity of calculations using Monte Carlo methods, uncertainty propagation is needed for depletion calculations.