Statistics and Probability Commons^™

A Note On The Positive Solutions Of An Inhomogeneous Elliptic Equation On Rn, Yinbin Deng, Yi Li, Fen Yang

Mathematics and Statistics Faculty Publications

This paper is contributed to the elliptic equation

(0.1) Δu+K(|x|)up+μf(|x|)=0,

where p>1, x∈Rn, n⩾3, and μ⩾0 is a constant. We study the structure of positive radial solutions of (0.1) and obtain the uniqueness of solution decaying faster than r−m at ∞ if μ is small enough under some assumptions on K and f, where m is the slow decay rate.

Periodic Traveling Waves In Sirs Endemic Models, Tong Li, Yi Li, Herbert W. Hethcote

Mathematics and Statistics Faculty Publications

Mathematical models are used to determine if infection wave fronts could occur by traveling geographically in a loop around a region or continent. These infection wave fronts arise by Hopf bifurcation for some spatial models for infectious disease transmission with distributed-contacts. Periodic traveling waves are shown to exist for the spatial analog of the SIRS endemic model, in which the temporary immunity is described by a delay, but they do not exist in a similar spatial SIRS endemic model without a delay. Specifically, we found that the ratio of the delay ω in the recovered class and the average infectious …

Extending The Skill Test For Disease Diagnosis, Shu-Chuan Lin, Paul H. Kvam, Jye-Chyi Lu

Department of Math & Statistics Faculty Publications

For diagnostic tests, we present an extension to the skill plot introduced by Briggs and Zaretski (Biometrics 2008; 64:250–261). The method is motivated by diagnostic measures for osteopetrosis in a study summarized by Hans et al. (The Lancet 1996; 348:511–514). Diagnostic test accuracy is typically defined using the area (or partial area) under the receiver operator characteristic (ROC) curve. If partial area is used, the resulting statistic can be highly subjective because the focus region of the ROC curve corresponds to a set of low false‐positive rates that are chosen by the experimenter. This paper introduces a more …

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 …

Statistical Methods In Microarray Data Analysis, Liping Huang

University of Kentucky Doctoral Dissertations

This dissertation includes three topics. First topic: Regularized estimation in the AFT model with high dimensional covariates. Second topic: A novel application of quantile regression for identification of biomarkers exemplified by equine cartilage microarray data. Third topic: Normalization and analysis of cDNA microarray using linear contrasts.

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. …

The Deterrent Effect Of Death Penalty Eligibility: Evidence From The Adoption Of Child Murder Eligibility Factors, Michael D. Frakes, Matthew Harding

Faculty Scholarship

We draw on within-state variations in the reach of capital punishment statutes between 1977 and 2004 to identify the deterrent effects associated with capital eligibility. Focusing on the most prevalent eligibility expansion, we estimate that the adoption of a child murder factor is associated with an approximately 20% reduction in the homicide rate of youth victims. Eligibility expansions may enhance deterrence by (1) paving the way for more executions and (2) providing prosecutors with greater leverage to secure enhanced non-capital sentences. While executions themselves are rare, this latter channel is likely to be triggered fairly regularly, providing a reasonable basis …

Comparative Analysis Of Forecasting Techniques With Intermittent Demand, John Deforest

Industrial Engineering Undergraduate Honors Theses

The world and life are filled with uncertainty. Statistics, and more specifically, forecasting techniques allow us to quantify uncertainty and make decisions based on that information. Many find forecasting advantageous in areas such as predicting consumer demand, stock prices, terrorist attacks, epidemiology, etc. This thesis will focus on forecasting intermittent demand, which is the study of analyzing sporadic demand. One application includes an airplane manufacturer’s sporadic or intermittent demand for spare parts in their distribution center. Since spare parts are not needed on any regular schedule, it is thought of as intermittent. Knowing the projected quantity for a given time …

"G" And "H" Control Charts And Risk Analysis In The Banking Industry, James L. Bossert

All-Inclusive List of Electronic Theses and Dissertations

This research investigates the utilization of a new control chart in the Banking industry to minimize financial risk. There are processes in the banking industry that do not lend themselves to traditional control chart applications. These processes tend to be high volume. High volume meaning over a million transactions a day and a requirement for high levels of accuracy characterize these processes. The research has looked at these processes and applied the "g" control charts to determine if they provide timely feedback to the banking industry. The value of this research will give the banking industry an opportunity to react …

Multi-Group Confirmatory Factor Analysis For Testing Measurement Invariance In Mixed Item Format Data, Kim H. Koh, Bruno D. Zumbo

Journal of Modern Applied Statistical Methods

This simulation study investigated the empirical Type I error rates of using the maximum likelihood estimation method and Pearson covariance matrix for multi-group confirmatory factor analysis (MGCFA) of full and strong measurement invariance hypotheses with mixed item format data that are ordinal in nature. The results indicate that mixed item formats and sample size combinations do not result in inflated empirical Type I error rates for rejecting the true measurement invariance hypotheses. Therefore, although the common methods are in a sense sub-optimal, they don’t lead to researchers claiming that measures are functioning differently across groups – i.e., a lack of …

Estimating Explanatory Power In A Simple Regression Model Via Smoothers, Rand R. Wilcox

Journal of Modern Applied Statistical Methods

Consider the regression model Y = γ(X) + ε , where γ(X) is some conditional measure of location associated with Y , given X. Let Υ̂ be some estimate of Y, given X, and let τ² (Y) be some measure of variation. Explanatory power is η² = τ² (Υ̂) /τ²(Y) . When γ(X) = β₀ + β₁X and τ²(Y) is the variance of Y , η² = ρ² , …

Data Mining Ceo Compensation, Susan M. Adams, Atul Gupta, Dominique M. Haughton, John D. Leeth

Journal of Modern Applied Statistical Methods

The need to pre-specify expected interactions between variables is an issue in multiple regression. Theoretical and practical considerations make it impossible to pre-specify all possible interactions. The functional form of the dependent variable on the predictors is unknown in many cases. Two ways are described in which the data mining technique Multivariate Adaptive Regression Splines (MARS) can be utilized: first, to obtain possible improvements in model specification, and second, to test for the robustness of findings from a regression analysis. An empirical illustration is provided to show how MARS can be used for both purposes.

Least Squares Percentage Regression, Chris Tofallis

Journal of Modern Applied Statistical Methods

In prediction, the percentage error is often felt to be more meaningful than the absolute error. We therefore extend the method of least squares to deal with percentage errors, for both simple and multiple regression. Exact expressions are derived for the coefficients, and we show how such models can be estimated using standard software. When the relative error is normally distributed, least squares percentage regression is shown to provide maximum likelihood estimates. The multiplicative error model is linked to least squares percentage regression in the same way that the standard additive error model is linked to ordinary least squares regression.

Application Of Dynamic Poisson Models To Japanese Cancer Mortality Data, Shuichi Midorikawa, Etsuo Miyaoka, Bruce Smith

Journal of Modern Applied Statistical Methods

A dynamic Poisson model is used with a Bayesian approach to modeling to predict cancer mortality. The complexity of the posterior distribution prohibits direct evaluation of the posterior, and so parameters are estimated by using a Markov Chain Monte Carlo method. The model is applied to analyze lung and stomach cancer data which have been collected in Japan.

A Randomization Method To Control The Type I Error Rates In Best Subset Regression, Yasser A. Shehata, Paul White

Journal of Modern Applied Statistical Methods

A randomization method for the assessment of statistical significance for best subsets regression is given. The procedure takes into account the number of potential predictors and the inter-dependence between predictors. The approach corrects a non-trivial problem with Type I errors and can be used to assess individual variable significance.

Comparing Factor Loadings In Exploratory Factor Analysis: A New Randomization Test, W. Holmes Finch, Brian F. French

Journal of Modern Applied Statistical Methods

Factorial invariance testing requires a referent loading to be constrained equal across groups. This study introduces a randomization test for comparing group exploratory factor analysis loadings so as to identify an invariant referent. Results show that it maintains the Type I error rate while providing adequate power under most conditions.

Variance Estimation In Time Series Regression Models, Samir Safi

Journal of Modern Applied Statistical Methods

The effect of variance estimation of regression coefficients when disturbances are serially correlated in time series regression models is studied. Variance estimation enters into confidence interval estimation, hypotheses testing, spectrum estimation, and expressions for the estimated standard error of prediction. Using computer simulations, the robustness of various estimators, including Estimated Generalized Least Squares (EGLS) was considered. The estimates of variance of the coefficient estimators produced by computer packages were considered. Models were generated with a second order auto-correlated error structure, considering the robustness of estimators based upon misspecified order. Ordinary Least Squares (OLS) (order zero) estimates outperformed first order EGLS. …

Two Dimension Marginal Distributions Of Crossing Time And Renewal Numbers Related To Two-Stage Erlang Processes, Mir Ghulam Hyder Talpur, Iffat Zamir, M. Masoom Ali

Journal of Modern Applied Statistical Methods

The two dimensional marginal transform, probability density and cumulative probability distribution functions for the random variables T_ξN (time taken by servers during vacations), ξ_N (number of vacations taken by servers) and N_η (number of customers or units arriving in the system) are derived by taking combinations of these random variables. One random variable is controlled at one time to determine the effect of the other two random variables simultaneously.

Bootstrap Confidence Intervals And Coverage Probabilities Of Regression Parameter Estimates Using Trimmed Elemental Estimation, Matthew Hall, Matthew S. Mayo

Journal of Modern Applied Statistical Methods

Mayo and Gray introduced the leverage residual-weighted elemental (LRWE) classification of regression estimators and a new method of estimation called trimmed elemental estimation (TEE), showing the efficiency and robustness of TEE point estimates. Using bootstrap methods, properties of various trimmed elemental estimator interval estimates to allow for inference are examined and estimates with ordinary least squares (OLS) and least sum of absolute values (LAV) are compared. Confidence intervals and coverage probabilities for the estimators using a variety of error distributions, sample sizes, and number of parameters are examined. To reduce computational intensity, randomly selecting elemental subsets to calculate the parameter …

Robust Predictive Inference For Multivariate Linear Models With Elliptically Contoured Distribution Using Bayesian, Classical And Structural Approaches, B. M. Golam Kibria

Journal of Modern Applied Statistical Methods

Predictive distributions of future response and future regression matrices under multivariate elliptically contoured distributions are discussed. Under the elliptically contoured response assumptions, these are identical to those obtained under matric normal or matric-t errors using structural, Bayesian with improper prior, or classical approaches. This gives inference robustness with respect to departure from the reference case of independent sampling from the matric normal or matric t to multivariate elliptically contoured distributions. The importance of the predictive distribution for skewed elliptical models is indicated; the elliptically contoured distribution, as well as matric t distribution, have significant applications in statistical practices.

Delete And Revise Procedures For Two-Stage Short-Run Control Charts, Matthew E. Elam

Journal of Modern Applied Statistical Methods

This article investigates the effect different delete and revise procedures have on the performance of twostage short-run control charting methodology in the second stage of its two stage procedure. Five variables control chart combinations, six delete and revise procedures, and various out-of-control situations in both stages are considered.

A Methodology To Improve Pci Use In Industry, Milind A. Phadnis, Matthew E. Elam

Journal of Modern Applied Statistical Methods

This article presents the development of a methodology using decision trees to resolve issues in industry with using process capability indices (PCIs). The methodology forms the structure of a prototype decision support system (PDSS) for PCI selection, calculation, and interpretation. Download instructions for the PDSS are available at http://program.20m.com.

The Multinomial Regression Modeling Of The Cause-Of-Death Mortality Of The Oldest Old In The U.S., Dudley L. Poston Jr., Hosik Min

Journal of Modern Applied Statistical Methods

The statistical modeling of the causes of death of the oldest old (persons aged 80 and over) in the U.S. in 2001 was conducted in this article. Data were analyzed using a multinomial logistic regression model (MNLM) because multiple causes of death are coded on death certificates and the codes are nominal. The percentage distribution of the 10 major causes of death among the oldest old was first examined; we next estimated a multinomial logistic regression equation to predict the likelihood of elders dying of one of the causes of death compared to dying of an “other cause.” The independent …

Frequency Domain Modeling With Piecewise Constant Spectra, Erhard Reschenhofer

Journal of Modern Applied Statistical Methods

Using piecewise constant functions as models for the spectral density of the differenced log real U.S. GDP it was found that these models have the capacity to compete with the spectral densities implied by ARMA models. According to AIC and BIC the piecewise constant spectral densities are superior to ARMA.

Correlation Between The Sample Mean And Sample Variance, Ramalingam Shanmugam

Journal of Modern Applied Statistical Methods

This article obtains a general formula to find the correlation coefficient between the sample mean and variance. Several particular results for major non-normal distributions are extracted to help students in classroom, clients during statistical consulting service.

Size-Biased Generalized Negative Binomial Distribution, Khurshid Ahmad Mir

Journal of Modern Applied Statistical Methods

A size biased generalized negative binomial distribution (SBGNBD) is defined and a recurrence relationship for the moments of SBGNBD is established. The Bayes’ estimator for a parametric function of one parameter when two other parameters of a known size-biased generalized negative binomial distribution is derived. Prior information on one parameter is given by a beta distribution and the parameters in the prior distribution are assigned by computer using Monte Carlo and R-software.

Non-Parametric Quantile Selection For Extreme Distributions, Wan Zawiah Wan Zin, Abdul Aziz Jemain

Journal of Modern Applied Statistical Methods

The objective is to select the best non-parametric quantile estimation method for extreme distributions. This serves as a starting point for further research in quantile application such as in parameter estimation using LQ-moments method. Thirteen methods of non-parametric quantile estimation were applied on six types of extreme distributions and their efficiencies compared. Monte Carlo methods were used to generate the results, which showed that the method of Weighted Kernel estimator of Type 1 was more efficient than the other methods in many cases.

Analyzing Incomplete Categorical Data: Revisiting Maximum Likelihood Estimation (Mle) Procedure, Hoo Ling Ping, M. Ataharul Islam

Journal of Modern Applied Statistical Methods

Incomplete data poses formidable difficulties in the application of statistical techniques and requires special procedures to handle. The most common ways to solve this problem are by ignoring, truncating, censoring or collapsing those data, but these may lead to inappropriate conclusions because those data might contain important information. Most of the research for estimating cell probabilities involving incomplete categorical data is based on the EM algorithm. A likelihood approach is employed for estimating cell probabilities for missing values and makes comparisons between maximum likelihood estimation (MLE) and the EM algorithm. The MLE can provide almost the same estimates as that …

Adaptive Estimation Of Heteroscedastic Linear Regression Model Using Probability Weighted Moments, Faqir Muhammad, Muhammad Aslam, G.R. Pasha

Journal of Modern Applied Statistical Methods

An adaptive estimator is presented by using probability weighted moments as weights rather than conventional estimates of variances for unknown heteroscedastic errors while estimating a heteroscedastic linear regression model. Empirical studies of the data generated by simulations for normal, uniform, and logistically distributed error terms support our proposed estimator to be quite efficient, especially for small samples.