Statistical Theory Commons^™

Bayesian Hypothesis Testing Of Two Normal Samples Using Bootstrap Prior Technique, Oyebayo Ridwan Olaniran, Waheed Babatunde Yahya

Journal of Modern Applied Statistical Methods

The most important ingredient in Bayesian analysis is prior or prior distribution. A new prior determination method was developed under the framework of parametric empirical Bayes using bootstrap technique. By way of example, Bayesian estimations of the parameters of a normal distribution with unknown mean and unknown variance conditions were considered, as well as its application in comparing the means of two independent normal samples with several scenarios. A Monte Carlo study was conducted to illustrate the proposed procedure in estimation and hypothesis testing. Results from Monte Carlo studies showed that the bootstrap prior proposed is more efficient than the …

Bayesian Model For Detection Of Outliers In Linear Regression With Application To Longitudinal Data, Zahraa Al-Sharea

Graduate Theses and Dissertations

Outlier detection is one of the most important challenges with many present-day applications. Outliers can occur due to uncertainty in data generating mechanisms or due to an error in data recording/processing. Outliers can drastically change the study's results and make predictions less reliable. Detecting outliers in longitudinal studies is quite challenging because this kind of study is working with observations that change over time. Therefore, the same subject can produce an outlier at one point in time produce regular observations at all other time points. A Bayesian hierarchical modeling assigns parameters that can quantify whether each observation is an outlier …

Examination And Comparison Of The Performance Of Common Non-Parametric And Robust Regression Models, Gregory F. Malek

Electronic Theses and Dissertations

ABSTRACT

Examination and Comparison of the Performance of Common Non-Parametric and Robust Regression Models

By

Gregory Frank Malek

Stephen F. Austin State University, Masters in Statistics Program,

Nacogdoches, Texas, U.S.A.

[email protected]

This work investigated common alternatives to the least-squares regression method in the presence of non-normally distributed errors. An initial literature review identified a variety of alternative methods, including Theil Regression, Wilcoxon Regression, Iteratively Re-Weighted Least Squares, Bounded-Influence Regression, and Bootstrapping methods. These methods were evaluated using a simple simulated example data set, as well as various real data sets, including math proficiency data, Belgian telephone call data, and faculty …

Prediction Of Stress Increase In Unbonded Tendons Using Sparse Principal Component Analysis, Eric Mckinney

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

While internal and external unbonded tendons are widely utilized in concrete structures, the analytic solution for the increase in unbonded tendon stress, Δ���, is challenging due to the lack of bond between strand and concrete. Moreover, most analysis methods do not provide high correlation due to the limited available test data. In this thesis, Principal Component Analysis (PCA), and Sparse Principal Component Analysis (SPCA) are employed on different sets of candidate variables, amongst the material and sectional properties from the database compiled by Maguire et al. [18]. Predictions of Δ��� are made via Principal Component Regression models, and the method …

Gilmore Girls And Instagram: A Statistical Look At The Popularity Of The Television Show Through The Lens Of An Instagram Page, Brittany Simmons

Student Scholar Symposium Abstracts and Posters

After going on the Warner Brothers Tour in December of 2015, I created a Gilmore Girls Instagram account. This account, which started off as a way for me to create edits of the show and post my photos from the tour turned into something bigger than I ever could have imagined. In just over a year I have over 55,000 followers. I post content including revival news, merchandise, and edits of the show that have been featured in Entertainment Weekly, Bustle, E! News, People Magazine, Yahoo News, & GilmoreNews.

I created a dataset of qualitative and quantitative outcomes from my …

Using Multiple Imputation To Address Missing Values Of Hierarchical Data, Yujia Zhang, Sara Crawford, Sheree Boulet, Michael Monsour, Bruce Cohen, Patricia Mckane, Karen Freeman

Journal of Modern Applied Statistical Methods

Missing data may be a concern for data analysis. If it has a hierarchical or nested structure, the SUDAAN package can be used for multiple imputation. This is illustrated with birth certificate data that was linked to the Centers for Disease Control and Prevention’s National Assisted Reproductive Technology Surveillance System database. The Cox-Iannacchione weighted sequential hot deck method was used to conduct multiple imputation for missing/unknown values of covariates in a logistic model.

Selection Of Statistical Software For Data Scientists And Teachers, Ceyhun Ozgur, Min Dou, Yang Li, Grace Rogers

Journal of Modern Applied Statistical Methods

The need for analysts with expertise in big data software is becoming more apparent in today’s society. Unfortunately, the demand for these analysts far exceeds the number available. A potential way to combat this shortage is to identify the software sought by employers and to align this with the software taught by universities. This paper will examine multiple data analysis software – Excel add-ins, SPSS, SAS, Minitab, and R – and it will outline the cost, training, statistical methods/tests/uses, and specific uses within industry for each of these software. It will further explain implications for universities and students.

An Unbiased Estimator Of The Greatest Lower Bound, Nol Bendermacher

Journal of Modern Applied Statistical Methods

The greatest lower bound to the reliability of a test, based on a single administration, is the Greatest Lower Bound (GLB). However the estimate is seriously biased. An algorithm is described that corrects this bias.

Monte Carlo Study Of Some Classification-Based Ridge Parameter Estimators, Adewale Folaranmi Lukman, Kayode Ayinde, Adegoke S. Ajiboye

Journal of Modern Applied Statistical Methods

Ridge estimator in linear regression model requires a ridge parameter, K, of which many have been proposed. In this study, estimators based on Dorugade (2014) and Adnan et al. (2014) were classified into different forms and various types using the idea of Lukman and Ayinde (2015). Some new ridge estimators were proposed. Results shows that the proposed estimators based on Adnan et al. (2014) perform generally better than the existing ones.

Multivariate Multilevel Modeling Of Age Related Diseases, Kapuruge N. O. Ranathunga, Roshini Sooriyarachchi

Journal of Modern Applied Statistical Methods

The emerging role of modeling multivariate multilevel data in the context of analyzing the risk factors are examined for the severity of cardiovascular disease diabetes, and chronic respiratory conditions. The modeling phase results leads to some important interaction terms between blood glucose, blood pressure, obesity, smoking and alcohol to the mortality rates.

Analysis Of Robust Parameter Designs, Tak K. Mak, Fassil Nebebe

Journal of Modern Applied Statistical Methods

The analysis of robust parameter design is discussed via a model incorporating mean-variance relationship which, when ignored as in the classical regression approach, can be problematic. The model is also capable of alleviating the difficulties of the regression approach in the search of the minimum variance occurring region.

Distribution Fits For Various Parameters In The Florida Public Hurricane Loss Model, Victoria Oxenyuk, Sneh Gulati, B. M. Golam Kibria, Shahid Hamid

Journal of Modern Applied Statistical Methods

The purpose of this study is to re-analyze the atmospheric science component of the Florida Public Hurricane Loss Model v. 5.0, in order to investigate if the distributional fits used for the model parameters could be improved upon. We consider alternate fits for annual hurricane occurrence, radius of maximum winds and the pressure profile parameter.

Stochastic Model For Cancer Cell Growth Through Single Forward Mutation, Jayabharathiraj Jayabalan

Journal of Modern Applied Statistical Methods

A stochastic model for cancer cell growth in any organ is presented, based on a single forward mutation. Cell growth is explained in a one-dimensional stochastic model, and statistical measures for the variable representing the number of malignant cells are derived. A numerical study is conducted to observe the behavior of the model.

Genetic Algorithms For Cross-Calibration Of Categorical Data, Suja M. Aboukhamseen, Rym A. M'Hallah

Journal of Modern Applied Statistical Methods

The probabilistic problem of cross-calibration of two categorical variables is addressed. A probabilistic forecast of the categorical variables is obtained based on a sample of observed data. This forecast is the output of a genetic algorithm based approach, which makes no assumption on the type of relationship between the two variables and applies a scoring rule to assess the fitness of the chromosomes. It converges to a good-quality point probability forecast of the joint distribution of the two variables. The proposed approach is applied both at stationary points in time and across time. Its performance is enhanced when additional sampled …

A Schmid-Leiman-Based Transformation Resulting In Perfect Inter-Correlations Of Three Types Of Factor Score Predictors, André Beauducel

Journal of Modern Applied Statistical Methods

Factor score predictors are computed when individual factor scores are of interest. Conditions for a perfect inter-correlation of the best linear factor score predictor, the best linear conditionally unbiased predictor, and the determinant best linear correlation-preserving predictor are presented. A transformation resulting in perfect correlations of the three predictors is proposed.

Errors In A Program For Approximating Confidence Intervals, Andrew V. Frane

Journal of Modern Applied Statistical Methods

An SPSS script previously presented in this journal contained nontrivial flaws. The script should not be used as written. A call is renewed for validation of new software.

An Empirical Comparison Between Robust Estimation And Robust Optimization To Mean-Variance Portfolio, Epha Diana Supandi, Dedi Rosadi, Abdurakhman

Journal of Modern Applied Statistical Methods

Mean-variance portfolios constructed using the sample mean and covariance matrix of asset returns perform poorly out-of-sample due to estimation error. Recently, there are two approaches designed to reduce the effect of estimation error: robust statistics and robust optimization. Two different robust portfolios were examined by assessing the out-of-sample performance and the stability of optimal portfolio compositions. The performance of the proposed robust portfolios was compared to classical portfolios via expected return, risk, and Sharpe Ratio. The aim is to shed light on the debate concerning the importance of the estimation error and weights stability in the portfolio allocation problem, and …

Guidelines For Generating Right-Censored Outcomes From A Cox Model Extended To Accommodate Time-Varying Covariates, Maria E. Montez-Rath, Kristopher Kapphahn, Maya B. Mathur, Aya A. Mitani, David J. Hendry, Manisha Desai

Journal of Modern Applied Statistical Methods

Simulating studies with right-censored outcomes as functions of time-varying covariates is discussed. Guidelines on the use of an algorithm developed by Zhou and implemented by Hendry are provided. Through simulation studies, the sensitivity of the method to user inputs is considered.

A Reinterpretation And Extension Of Mcnemar’S Test, Chauncey M. Dayton

Journal of Modern Applied Statistical Methods

The McNemar test is extended to multiple groups based on a latent class model incorporating classes representing consistent responders and a single latent error rate. The method is illustrated with data from a CDC survey of immunizations for flu and pneumonia for which a part-heterogeneous model is selected for interpretation.

In Response To Frane, "Errors In A Program For Approximating Confidence Intervals", David A. Walker

Journal of Modern Applied Statistical Methods

A rebuttal to Frane's letter to the Editor in this issue.

Experiment-Wise Type I Error Rates In Nested (Hierarchical) Study Designs, Jack Sawilowsky, Barry Markman

Journal of Modern Applied Statistical Methods

When conducting a statistical test one of the initial risks that must be considered is a Type I error, also known as a false positive. The Type I error rate is set by nominal alpha, assuming all underlying conditions of the statistic are met. Experiment-wise Type I error inflation occurs when multiple tests are conducted overall for a single experiment. There is a growing trend in the social and behavioral sciences utilizing nested designs. A Monte Carlo study was conducted using a two-layer design. Five theoretical distributions and four real datasets taken from Micceri (1989) were used, each with five …

Control Charts For Mean For Non-Normally Correlated Data, J. R. Singh, Ab Latif Dar

Journal of Modern Applied Statistical Methods

Traditionally, quality control methodology is based on the assumption that serially-generated data are independent and normally distributed. On the basis of these assumptions the operating characteristic (OC) function of the control chart is derived after setting the control limits. But in practice, many of the basic industrial variables do not satisfy both the assumptions and hence one may doubt the validity of the inferences drawn from the control charts. In this paper the power of the control chart for the mean is examined when both the assumptions of independence and normality are not tenable. The OC function is calculated and …

Multivariate Rank Outlyingness And Correlation Effects, Olusola Samuel Makinde

Journal of Modern Applied Statistical Methods

The effect of correlation on multivariate rank outlyingness, a result of deviation of multivariate rank functions from property of spherical symmetry, is examined. Possible affine invariant versions of this multivariate rank are surveyed, and outlyingness of affine invariant and non-invariant spatial rank functions under general affine transformation are compared.

A Comparison Of Different Methods Of Zero-Inflated Data Analysis And An Application In Health Surveys, Si Yang, Lisa L. Harlow, Gavino Puggioni, Colleen A. Redding

Journal of Modern Applied Statistical Methods

The performance of several models under different conditions of zero-inflation and dispersion are evaluated. Results from simulated and real data showed that the zero-altered or zero-inflated negative binomial model were preferred over others (e.g., ordinary least-squares regression with log-transformed outcome, Poisson model) when data have excessive zeros and over-dispersion.

Test Statistics For The Comparison Of Means For Two Samples That Include Both Paired And Independent Observations, Ben Derrick, Bethan Russ, Deirdre Toher, Paul White

Journal of Modern Applied Statistical Methods

Standard approaches for analyzing the difference in two means, where partially overlapping samples are present, are less than desirable. Here are introduced two test statistics, making reference to the t-distribution. It is shown that these test statistics are Type I error robust, and more powerful than standard tests.

Graphical Log-Linear Models: Fundamental Concepts And Applications, Niharika Gauraha

Journal of Modern Applied Statistical Methods

A comprehensive study of graphical log-linear models for contingency tables is presented. High-dimensional contingency tables arise in many areas. Analysis of contingency tables involving several factors or categorical variables is very hard. To determine interactions among various factors, graphical and decomposable log-linear models are preferred. Connections between the conditional independence in probability and graphs are explored, followed with illustrations to describe how graphical log-linear model are useful to interpret the conditional independences between factors. The problem of estimation and model selection in decomposable models is discussed.

Robustness And Power Comparison Of The Mood-Westenberg And Siegel-Tukey Tests, Linda C. Lowenstein, Shlomo S. Sawilowsky

Journal of Modern Applied Statistical Methods

The Mood-Westenberg and Siegel-Tukey tests were examined to determine their robustness with respect to Type-I error for detecting variance changes when their assumptions of equal means were slightly violated, a condition that approaches the Behrens-Fisher problem. Monte Carlo methods were used via 34,606 variations of sample sizes, α levels, distributions/data sets, treatments modeled as a change in scale, and treatments modeled as a shift in means. The Siegel-Tukey was the more robust, and was able to handle a more diverse set of conditions.

A New Estimator For The Pickands Dependence Function, Marta Ferreira

Journal of Modern Applied Statistical Methods

The Pickands dependence function characterizes an extreme value copula, a useful tool in the modeling of multivariate extremes. A new estimator is presented along with its convergence properties and performance through simulation.

Effective Estimation Strategy Of Finite Population Variance Using Multi-Auxiliary Variables In Double Sampling, Reba Maji, G. N. Singh, Arnab Bandyopadhyay

Journal of Modern Applied Statistical Methods

Estimation of population variance in two-phase (double) sampling is considered using information on multiple auxiliary variables. An unbiased estimator is proposed and its properties are studied under two different structures. The superiority of the suggested estimator over some contemporary estimators of population variance was established through empirical studies from a natural and an artificially generated dataset.

Vol. 16, No. 1 (Full Issue), Jmasm Editors

Journal of Modern Applied Statistical Methods

.