Physical Sciences and Mathematics Commons^™

Statistical Analysis Of Linear Analog Circuits Using Gaussian Message Passing In Factor Graphs, Miti Phadnis

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

This thesis introduces a novel application of factor graphs to the domain of analog circuits. It proposes a technique of leveraging factor graphs for performing statistical yield analysis of analog circuits that is much faster than the standard Monte Carlo/Simulation Program With Integrated Circuit Emphasis (SPICE) simulation techniques. We have designed a tool chain to model an analog circuit and its corresponding factor graph and then use a Gaussian message passing approach along the edges of the graph for yield calculation. The tool is also capable of estimating unknown parameters of the circuit given known output statistics through backward message …

Theory Of Planned Behavior Model Fit Using Atod Prevention Program Data, Ying Jin

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

This report is to test the Theory of Planned Behavior (TpB) model fit using the data collected from the ATOD Prevention Program conducted by the Operation Snowball Program from year 2004 to 2007 in Naperville, Illinois. Measurement Model and Structural Equation Modeling are used as principal modeling methods to test internal consistency of assigned measures for each construct and the dependency between constructs respectively. The results show that the ATOD Prevention Program data does not fit the TpB model perfectly. Extra paths should be added to the original theoretical model in order to obtain a satisfactory model fit.

The Effectiveness Of Institutional Intervention At Minimizing Demographic Inertia And Improving The Representation Of Women Faculty In Higher Education, Amanda V. Bakian

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Across all science and engineering disciplines, women currently receive 46% of the doctoral degrees granted annually. Despite gains in doctoral degrees earned by women over the past 30 years, they remain under-represented among full time tenured /tenure-track science and engineering faculty compared to their presence among degree recipients. The inequality gap is widest among full professors at 4 year research universities where women account for a mere 16% of all S&E full time full professors. Multiple hypothesis have been proposed to account for women's under-representation relative to men including those based on human capital/economic theory, feminist theory, innate biological gender …

Investigations Of Variable Importance Measures Within Random Forests, Andrew C. Merrill

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Random Forests (RF) (Breiman 2001; Breiman and Cutler 2004) is a completely nonparametric statistical learning procedure that may be used for regression analysis and. A feature of RF that is drawing a lot of attention is the novel algorithm that is used to evaluate the relative importance of the predictor/explanatory variables. Other machine learning algorithms for regression and classification, such as support vector machines and artificial neural networks (Hastie et al. 2009), exhibit high predictive accuracy but provide little insight into predictive power of individual variables. In contrast, the permutation algorithm of RF has already established a track record for …

Logistic Models With Missing Categorical Covariates, Jeremiah Rounds

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

We present an EM based solution to missing categorical covariates in Binomial models with logit links using an assumption that experimental units are drawn from a Multinomial population of infinite size. We further address the problem of separation of points inducing large variances on parameter estimates by the use of a novel score-modification based on Firth's bias-reduction score-modification. We simulate to address questions about estimate bias, distribution, and appropriate parameter coverage by Wald intervals.

Approaches To Promote Active, Conceptual Learning In A Pedagogically Hybrid Introductory Statistics Course, Brittany L. Allred

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Statistics education is an active area of research where discovery-based learning is becoming more prominent. This project reviews how the USEI and GAISE recommendations can be implemented in the statistics classroom. Further, the project describes the creation of a library of classroom materials for an introductory statistics course. The results are also discussed from implementing various library materials, along with the student response to hands-on learning techniques.

The Statistical Evaluation Of Collective Risk Models With Various Distributional Assumptions, Chang Li

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Our thesis includes 2 sections. In section 1, we mainly discuss the distribution function and the empirical simulation of the total claims amount for a portfolio with multiple distributional assumptions, under the collective risk model for a single period.

In section 2, we discuss the collective risk model for an extended period, which mainly deals with the amount of surplus over several periods of time. Let Un denote the surplus at time n, then we interested in the probability of ruin, that is, the probability that there exists an n such that Un< O. In this section, we put forward an algorithm for the probability of ruin (both theoretical and empirical) using this model, with various distributional assumptions, and compare the theoretical and empirical results.

The reference of this project is Actuarial …

Probe-Level Statistical Models For Differential Expression Of Genes In Bovine Nt Studies, Jason L. Bell

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

A brief introduction of microarray technology and its uses is given. This technology is commonly used in agricultural research, including research in nuclear transfer, which motivated this study. There are 3 classes of statistical models compared: probeset-level, weighted probeset-level and probe-level.

Different statistical mod els are compared on 3 spike-in experiments to assess the relative performance in identifying differentially expressed genes . A novel nested factorial model was found to outperform all other models compared in this study in one spike-in experiment, and was found to be competitive in its performance relative to the other models on the other spike-in …

Comparison Of Random Forests And Cforest: Variable Importance Measures And Prediction Accuracies, Rong Xia

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Random forests are ensembles of trees that give accurate predictions for regression, classification and clustering problems. The CART tree, the base learn er employed by random forests, has been criticized because of bias in the selection of splitting variables. The performance of random forests is suspect due to this criticism. A new implementation of random forests, Cforest, which is claimed to outperform random forests in both predictive power and variable importance measures , was developed based on Ctree, an implementation of conditional inference trees.

We address the underlying mechanism of random forests and Cforest in this report. Comparison of random …

An Investigation Of The Ends Of Finitely Generated Groups, Daniel T. Murphree

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Geometric group theory is a relatively new branch of mathematics, studied as a distinct area since the 1990's. It explores invariant properties of groups based on group actions defined on topological or geometrical spaces. One of the pioneering works in geometric group theory is the article "Topological Methods in Group Theory" by Peter Scott and Terry Wall, written in 1977. This article was an overview of revised notes from an advanced course give in Liverpool in the same year. This report is an attempt to make these notes more accessible to lower level graduate students in the fields of topology …

Applications Of Modern Classification Methods To Predict Occurrences Of Rare And Invasive Plant Species, Stacey Reynolds

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Statistical classification methods are among the most widely used statistical procedures in ecology. Applications include vegetation mapping by remote sensing (Steele 2000), discrimination of subspecies using morphological measurements (Fisher 1936, 1938; Conner and Schenk 2003), and species distribution modelling (Guisan and Zimmerman 2000). Examples of the last application abound in the ecological literature and include predicting the distribution or characteristics of plant species (see, e.g., Austin et al. 1990), predicting presence and absence aquatic biota in streams (Hawkins et al. 2000), and habit at relationships of terrestrial animal species (Welch and MacMahon 2005). Over the last 20 years two mainstays …

Cow Insurance Vs. Human Insurance: A Comparison Study, Corlan Fawcett

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

This paper will compare the premiums to insure cows and humans for n-year term, n-year endowment, and whole life insurance. It will go into more depth for whole life premium s that would be charged, such that the insurance company is 95 percent confident that they will not lose money. In particular the paper will look at how the variance affects the premiums for an insurance company insuring 100 cows.

Evaluating The Effectiveness Of A Probability Applet, Chelsa Green Tuddenham

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

The purpose of this project was to develop and examine a computerized interactive probability applet designed to teach basic concepts of probability. The applet design was intended to be a user-friendly way to encourage understanding of basic probability concepts such as, rules of probability, probabilities of mutually exclusive events, and the multiplication rule. Thirty participants were recruited from Utah State University and were asked to participate in the evaluation process of the probability applet. The evaluation of the probability applet consisted of three parts: First, a group of participants were video recorded and asked to voice any and all thoughts …

Comparison Of Exact Methods For Analyzing Family-Based Samples, Abbie Lundgreen

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Family-based association tests are used to identify genes that increase the risk of developing a disease, while controlling for spurious associations caused by population structure. The exact family-based association test, exact FBAT, is a unified approach which can be app lied to tests of different genetic models, sampling designs, null hypotheses , and missing parental information.

The purpose of this report is to compare the power of the exact FBAT with two other tests, exact conditional logistic regression (CLR) and the exact trend test for clustered data (QEM). Pedigrees of sibships were simulated based upon a variety of different parameters, …