Examination And Comparison Of The Performance Of Common Non-Parametric And Robust Regression Models, 2017 Stephen F Austin State University

#### 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.

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 ...

Prediction Of Stress Increase In Unbonded Tendons Using Sparse Principal Component Analysis, 2017 Utah State University

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

*All Graduate Plan B and other Reports*

While internal and external unbonded tendons are widely utilized in concrete structures, the analytic solution for the increase in unbonded tendon stress, Δ𝑓𝑝𝑠, 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 Δ𝑓𝑝𝑠 are made via Principal Component Regression models, and the method ...

Propensity Score Analysis With Matching Weights, 2017 Cleveland Clinic

#### Propensity Score Analysis With Matching Weights, Liang Li

*Liang Li*

The propensity score analysis is one of the most widely used methods for studying the causal treatment effect in observational studies. This paper studies treatment effect estimation with the method of matching weights. This method resembles propensity score matching but offers a number of new features including efficient estimation, rigorous variance calculation, simple asymptotics, statistical tests of balance, clearly identified target population with optimal sampling property, and no need for choosing matching algorithm and caliper size. In addition, we propose the mirror histogram as a useful tool for graphically displaying balance. The method also shares some features of the inverse ...

Gilmore Girls And Instagram: A Statistical Look At The Popularity Of The Television Show Through The Lens Of An Instagram Page, 2017 Chapman University

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

*Student Research Day 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 ...

Random Regression Models Based On The Elliptically Contoured Distribution Assumptions With Applications To Longitudinal Data, 2017 University of Alabama at Birmingham

#### Random Regression Models Based On The Elliptically Contoured Distribution Assumptions With Applications To Longitudinal Data, Alfred A. Bartolucci, Shimin Zheng, Sejong Bae, Karan P. Singh

*Shimin Zheng*

We generalize Lyles et al.’s (2000) random regression models for longitudinal data, accounting for both undetectable values and informative drop-outs in the distribution assumptions. Our models are constructed on the generalized multivariate theory which is based on the Elliptically Contoured Distribution (ECD). The estimation of the fixed parameters in the random regression models are invariant under the normal or the ECD assumptions. For the Human Immunodeficiency Virus Epidemiology Research Study data, ECD models fit the data better than classical normal models according to the Akaike (1974) Information Criterion. We also note that both univariate distributions of the random intercept ...

Vol. 16, No. 1 (Full Issue), 2017 Wayne State University

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

*Journal of Modern Applied Statistical Methods*

.

A New Estimator Based On Auxiliary Information Through Quantitative Randomized Response Techniques, 2017 Hacettepe University

#### A New Estimator Based On Auxiliary Information Through Quantitative Randomized Response Techniques, Nilgün Özgül, Hülya Çıngı

*Journal of Modern Applied Statistical Methods*

An exponential-type estimator is developed for the population mean of the sensitive study variable based on various Randomized Response Techniques (RRT) using a non-sensitive auxiliary variable. The mean squared error (MSE) of the proposed estimator is derived for generalized RRT models. The proposed estimator is compared with competitors in a simulation study and an application. The proposed estimator is found to be more efficient using a non-sensitive auxiliary variable.

A Comparison Of Depth Functions In Maximal Depth Classification Rules, 2017 Federal University of Technology, Akure, Nigeria

#### A Comparison Of Depth Functions In Maximal Depth Classification Rules, Olusola Samuel Makinde, Adeyinka Damilare Adewumi

*Journal of Modern Applied Statistical Methods*

Data depth has been described as alternative to some parametric approaches in analyzing many multivariate data. Many depth functions have emerged over two decades and studied in literature. In this study, a nonparametric approach to classification based on notions of different data depth functions is considered and some properties of these methods are studied. The performance of different depth functions in maximal depth classifiers is investigated using simulation and real data with application to agricultural industry.

An Extended Weighted Exponential Distribution, 2017 Department of Statistics, Faculty of Mathematical Sciences, Vali-e-Asr University of Rafsanjan, Rafsanjan, Iran

#### An Extended Weighted Exponential Distribution, Abbas Mahdavi, Leila Jabari

*Journal of Modern Applied Statistical Methods*

A new class of weighted distributions is proposed by incorporating an extended exponential distribution in Azzalini’s (1985) method. Several statistics and reliability properties of this new class of distribution are obtained. Maximum likelihood estimators of the unknown parameters cannot be obtained in explicit forms; they have to be obtained by solving some numerical methods. Two data sets are analyzed for illustrative purposes, and show that the proposed model can be used effectively in analyzing real data.

A Note On Determination Of Sample Size From The Perspective Of Six Sigma Quality, 2017 Amrita Vishwa Vidyapeetham, Amrita University

#### A Note On Determination Of Sample Size From The Perspective Of Six Sigma Quality, Joghee Ravichandran

*Journal of Modern Applied Statistical Methods*

In most empirical studies (clinical, network modeling, and survey-based and aeronautical studies, etc.), sample observations are drawn from population to analyze and draw inferences about the population. Such analysis is done with reference to a measurable quality characteristic of a product or process of interest. However, fixing a sample size is an important task that has to be decided by the experimenter. One of the means in deciding an appropriate sample size is the fixation of error limit and the associated confidence level. This implies that the analysis based on the sample used must guarantee the prefixed error and confidence ...

A Schmid-Leiman-Based Transformation Resulting In Perfect Inter-Correlations Of Three Types Of Factor Score Predictors, 2017 Institute of Psychology, University of Bonn, Germany

#### 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.

Robust Ancova: Confidence Intervals That Have Some Specified Simultaneous Probability Coverage When There Is Curvature And Two Covariates, 2017 University of Southern California

#### Robust Ancova: Confidence Intervals That Have Some Specified Simultaneous Probability Coverage When There Is Curvature And Two Covariates, Rand Wilcox

*Journal of Modern Applied Statistical Methods*

Consider the commonly occurring situation where the goal is to compare two independent groups and there are two covariates. Let Mj(X) be some conditional measure of location for the jth group associated with some random variable Y given X = (X1, X2). The goal is to H0: M1(X) = M2(X) for each X Ω in a manner that controls the probability of one or more Type I errors. An extant technique (method M1 here) addresses this goal without making any parametric assumption about Mj(X). However, a practical concern is that it does not provide enough detail regarding where ...

Experiment-Wise Type I Error Rates In Nested (Hierarchical) Study Designs, 2017 Citigroup

#### 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 ...

Plant Leaf Image Detection Method Using A Midpoint Circle Algorithm For Shape-Based Feature Extraction, 2017 K.L.N. College of Engineering, Tamil Nadu, India

#### Plant Leaf Image Detection Method Using A Midpoint Circle Algorithm For Shape-Based Feature Extraction, B. Vijaya Lakshmi, V. Mohan

*Journal of Modern Applied Statistical Methods*

Shape-based feature extraction in content-based image retrieval is an important research area at present. An algorithm is presented, based on shape features, to enhance the set of features useful in a leaf identification system.

Monte Carlo Study Of Some Classification-Based Ridge Parameter Estimators, 2017 Ladoke Akintola University of Technology

#### 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, 2017 University of Colombo, Sri Lanka

#### 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.

Stochastic Model For Cancer Cell Growth Through Single Forward Mutation, 2017 Pondicherry University

#### 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.

A Comparison Of Different Methods Of Zero-Inflated Data Analysis And An Application In Health Surveys, 2017 University of Rhode Island

#### 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.

Robustness And Power Comparison Of The Mood-Westenberg And Siegel-Tukey Tests, 2017 Wasthington University in St. Louis

#### 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.

Multivariate Rank Outlyingness And Correlation Effects, 2017 Department of Statistics, Federal University of Technology, P.M.B. 704, Akure, Nigeria

#### 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.