Statistics and Probability Commons^™

Population Mean Estimation With Sub Sampling The Non-Respondents Using Two Phase Sampling, Sunil Kumar, M Viswanathaiah

Journal of Modern Applied Statistical Methods

The problem of non-response in double (or two phase) sampling is dealt with combined ratio, product and regression estimators. Expressions of bias and MSE for these estimators are obtained. Comparisons of a proposed strategy with a usual unbiased estimator and other estimators are carried out and results obtained are illustrated numerically using an empirical sample.

Estimation And Testing In Type-Ii Generalized Half Logistic Distribution, R R. L. Kantam, V Ramakrishna, M S. Ravikumar

Journal of Modern Applied Statistical Methods

A generalization of the Half Logistic Distribution is developed through exponentiation of its survival function and named the Type II Generalized Half Logistic Distribution (GHLD). The distributional characteristics are presented and estimation of its parameters using maximum likelihood and modified maximum likelihood methods is studied with comparisons. Discrimination between Type II GHLD and exponential distribution in pairs is conducted via likelihood ratio criterion.

A Compound Of Geeta Distribution With Generalized Beta Distribution, Adil Rashid, T R. Jan

Journal of Modern Applied Statistical Methods

A compound of Geeta distribution with Generalized Beta distribution (GBD) is obtained and the compound is specialized for different values of β. The first order factorial moments of some special compound distributions are also obtained. A chronological overview of recent developments in the compounding of distributions is provided in the introduction.

Hierarchical Clustering With Simple Matching And Joint Entropy Dissimilarity Measure, A Mete ÇilingtüRk, ÖZlem ErgüT

Journal of Modern Applied Statistical Methods

Conventional clustering algorithms are restricted for use with data containing ratio or interval scale variables; hence, distances are used. As social studies require merely categorical data, the literature is enriched with more complicated clustering techniques and algorithms of categorical data. These techniques are based on similarity or dissimilarity matrices. The algorithms are using density based or pattern based approaches. A probabilistic nature to similarity structure is proposed. The entropy dissimilarity measure has comparable results with simple matching dissimilarity at hierarchical clustering. It overcomes dimension increase through binarization of the categorical data. This approach is also functional with the clustering methods, …

An Exploratory Graphical Method For Identifying Associations In R X C Contingency Tables, Martin L. Lesser, Meredith B. Akerman

Journal of Modern Applied Statistical Methods

On finding a significant association between rows and columns of an r x c contingency table, the next step is to study the nature of the association in more detail. The use of a scree plot to visualize the largest contributions to Χ² among all cells in the table in order to determine the nature of the association in more detail is proposed.

Separate Ratio-Type Estimators Of Population Mean In Stratified Random Sampling, Rajesh Tailor, Hilal A. Lone

Journal of Modern Applied Statistical Methods

Separate ratio-type estimators for population mean with their properties are considered. Some separate ratio-type estimators for population mean using known parameters of auxiliary variate are proposed. The bias and mean squared error of the proposed estimators are obtained up to the first degree of approximation. It is shown that the proposed estimators are more efficient than unbiased estimators in stratified random sampling and usual separate ratio estimators under certain obtained conditions. To judge the merits of the proposed estimators, an empirical study was conducted.

Evaluation Of Area Under The Constant Shape Bi-Weibull Roc Curve, Sudesh Pundir, R Amala

Journal of Modern Applied Statistical Methods

The Receiver Operating Characteristic (ROC) curve generated based on assuming a constant shape Bi-Weibull distribution is studied. In the context of ROC curve analysis, it is assumed that biomarker values from controls and cases follow some specific distribution and the accuracy is evaluated by using the ROC model developed from that specified distribution. This article assumes that the biomarker values from the two groups follow Weibull distributions with equal shape parameter and different scale parameters. The ROC model, area under the ROC curve (AUC), asymptotic and bootstrap confidence intervals for the AUC are derived. Theoretical results are validated by simulation …

Investigating The Feasibility Of Using Mplus In The Estimation Of Growth Mixture Models, Ming Li, Jeffrey R. Harring, George B. Macready

Journal of Modern Applied Statistical Methods

Hipp and Bauer (2006) investigated the issues of singularities and local maximum solutions within growth mixture models (GMMs) and made recommendations regarding the use of multiple starting values. Building on their work, this simulation study investigates the feasibility of estimating GMMs within Mplus as measured by convergence to proper, but local solutions.

What Residualizing Predictors In Regression Analyses Does (And What It Does Not Do), Lee H. Wurm, Sebastiano A. Fisicaro

Psychology Faculty Research Publications

Psycholinguists are making increasing use of regression analyses and mixed-effects modeling. In an attempt to deal with concerns about collinearity, a number of researchers orthogonalize predictor variables by residualizing (i.e., by regressing one predictor onto another, and using the residuals as a stand-in for the original predictor). In the current study, the effects of residualizing predictor variables are demonstrated and discussed using ordinary least-squares regression and mixed-effects models. Some of these effects are almost certainly not what the researcher intended and are probably highly undesirable. Most importantly, what residualizing does not do is change the result for the residualized variable, …

Comparing Partial Least Square Approaches In Gene-Or Region-Based Association Study For Multiple Quantitative Phenotypes, Zhongshang Yuan, Xiaoshuai Zhang, Fangyu Li, Jinghua Zhao, Fuzhong Xue

Human Biology Open Access Pre-Prints

On thinking quantitatively of complex diseases, there are at least three statistical strategies for association study: single SNP on single trait, gene-or region (with multiple SNPs) on single trait and on multiple traits. The third of which is the most general in dissecting the genetic mechanism underlying complex diseases underpinning multiple quantitative traits. Gene-or region association methods based on partial least square (PLS) approaches have been shown to have apparent power advantage. However, few attempts are developed for multiple quantitative phenotypes or traits underlying a condition or disease, and the performance of various PLS approaches used in association study for …

On The Joys Of Missing Data, Todd D. Little, Terrence D. Jorgensen, Kyle M. Lang, E. Whitney G. Moore

Kinesiology, Health and Sport Studies

We provide conceptual introductions to missingness mechanisms—missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR)—and state-of-the-art methods of handling missing data—full-information maximum likelihood (FIML) and multiple imputation (MI)—followed by a discussion of planned missing designs: multiform questionnaire protocols, two-method measurement models, and wave-missing longitudinal designs. We reviewed 80 articles of empirical studies published in the 2012 issues of the Journal of Pediatric Psychology to present a picture of how adequately missing data are currently handled in this field. To illustrate the benefits of utilizing MI or FIML and incorporating planned missingness into study designs, …

Adaptive Stochastic Systems: Estimation, Filtering, And Noise Attenuation, Araz Ryan Hashemi

Wayne State University Dissertations

This dissertation investigates problems arising in identification and control of stochastic systems. When the parameters determining the underlying systems are unknown and/or time varying, estimation and adaptive filter- ing are invoked to to identify parameters or to track time-varying systems. We begin by considering linear systems whose coefficients evolve as a slowly- varying Markov Chain. We propose three families of constant step-size (or gain size) algorithms for estimating and tracking the coefficient parameter: Least-Mean Squares (LMS), Sign-Regressor (SR), and Sign-Error (SE) algorithms.

The analysis is carried out in a multi-scale framework considering the relative size of the gain (rate of …

Planned Missing Data Designs & Small Sample Size: How Small Is Too Small?, Fan Jia, E. Whitney G. Moore, Richard Kinai, Kelly S. Crowe, Alexander M. Schoemann, Todd D. Little

Kinesiology, Health and Sport Studies

Utilizing planned missing data (PMD) designs (ex. 3-form surveys) enables researchers to ask participants fewer questions during the data collection process. An important question, however, is just how few participants are needed to effectively employ planned missing data designs in research studies. This paper explores this question by using simulated three-form planned missing data to assess analytic model convergence, parameter estimate bias, standard error bias, mean squared error (MSE), and relative efficiency (RE).Three models were examined: a one-time point, cross-sectional model with 3 constructs; a two-time point model with 3 constructs at each time point; and a three-time point, mediation …

Lost In Translation: Statistical Inference In Court, Erica Beecher-Monas

Law Faculty Research Publications

No abstract provided.

The Impact Of Nested Testing On Experiment-Wise Type I Error Rate, Jack Sawilowsky

Wayne State University Dissertations

When conducting a statistical test the initial risk 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 different …

Robust Regression Methods For Massively Decayed Intelligence Data, Akiva Joachim Lorenz

Wayne State University Dissertations

Homeland Security, sponsored by governmental initiatives, has become a vibrant academic research field. However, most efforts were placed with the recognition of threats (e.g. theory) and response options. Less effort was placed in the analysis of the collected data through statistical modeling. In a field that collects more than 20 terabyte of information per minute though diverse overt and covert means and indexes it for future research, understanding how different statistical models behave when it comes to massively decayed data is of vital importance.

Using Monte Carlo methods, three regression techniques (ordinary least squares, least-trimmed, and maximum likelihood) were tested …

A Monte Carlo Comparison Of Robust Manova Test Statistics, Holmes Finch, Brian French

Journal of Modern Applied Statistical Methods

Multivariate Analysis of Variance (MANOVA) is a popular statistical tool in the social sciences, allowing for the comparison of mean vectors across groups. MANOVA rests on three primary assumptions regarding the population: (a) multivariate normality, (b) equality of group population covariance matrices and (c) independence of errors. When these assumptions are violated, MANOVA does not perform well with respect to Type I error and power. There are several alternative test statistics that can be considered including robust statistics and the use of the structural equation modeling (SEM) framework. This simulation study focused on comparing the performance of the P test …

On Some Properties Of A Heterogeneous Transfer Function Involving Symmetric Saturated Linear (Satlins) With Hyperbolic Tangent (Tanh) Transfer Functions, Christopher Godwin Udomboso

Journal of Modern Applied Statistical Methods

For transfer functions to map the input layer of the statistical neural network model to the output layer perfectly, they must lie within bounds that characterize probability distributions. The heterogeneous transfer function, SATLINS_TANH, is established as a Probability Distribution Function (p.d.f), and its mean and variance are shown.

Distribution Of The Ratio Of Normal And Rice Random Variables, Nayereh B. Khoolenjani, Kavoos Khorshidian

Journal of Modern Applied Statistical Methods

The ratio of independent random variables arises in many applied problems. The distribution of the ratio |X/Y| is studied when X and Y are independent Normal and Rice random variables, respectively. Ratios of such random variables have extensive applications in the analysis of noises in communication systems. The exact forms of probability density function (PDF), cumulative distribution function (CDF) and the existing moments are derived in terms of several special functions. As a special case, the PDF and CDF of the ratio of independent standard Normal and Rayleigh random variables have been obtained. Tabulations of associated percentage points …

Front Matter, Jmasm Editors

Journal of Modern Applied Statistical Methods

No abstract provided.

Preliminary Testing For Normality: Is This A Good Practice?, H. J. Keselman, Abdul R. Othman, Rand R. Wilcox

Journal of Modern Applied Statistical Methods

Normality is a distributional requirement of classical test statistics. In order for the test statistic to provide valid results leading to sound and reliable conclusions this requirement must be satisfied. In the not too distant past, it was claimed that violations of normality would not likely jeopardize scientific findings (See Hsu & Feldt, 1969; Lunney, 1970). Recent revelations suggest otherwise (See e.g., Micceri, 1989; Keselman, Huberty, Lix et al., 1998; Erceg-Hurn, Wilcox, & Keselman, 2013; Wilcox and Keselman, 2003; Wilcox, 2012a, b). Unfortunately the data obtained in psychological investigations rarely, if ever, meet the requirement of normally distributed data (Micceri, …

Intrinsically Ties Adjusted Non-Parametric Method For The Analysis Of Two Sampled Data, G. U. Ebuh, I. C. A Oyeka

Journal of Modern Applied Statistical Methods

A non-parametric method for the analysis of two sample data is proposed that intrinsically and structurally adjusts the test statistic for the possible presence of tied observations between the sampled populations, thereby obviating the need to require the populations to be continuous. The populations may be measurements on as low as the ordinal scale, and need not be homogeneous. In cases where the null hypotheses are rejected, the test statistic enables the determination of which of the sampled populations is likely to be responsible for the rejection (a determination which the Wilcoxon Mann Whitney test cannot handle). The proposed method …

The Impact Of Continuity Violation On Anova And Alternative Methods, Björn Lantz

Journal of Modern Applied Statistical Methods

The normality assumption behind ANOVA and other parametric methods implies that response variables are measured on continuous scales. A simulation approach is used to explore the impact of continuity violation on the performance of statistical methods commonly used by applied researchers to compare locations across several groups.

The Single-Case Data Analysis Package: Analysing Single-Case Experiments With R Software, Isis Bulté, Patrick Onghena

Journal of Modern Applied Statistical Methods

The RcmdrPlugin.SCDA plug-in package is discussed. It integrates three R packages in the R commander interface: SCVA (for Single-Case Visual Analysis), SCRT (for Single-Case Randomization Tests), and SCMA (for Single-Case Meta-Analysis). This way the plug-in package covers three important steps in the analysis of single-case data.

A Comparison Between Biased And Unbiased Estimators In Ordinary Least Squares Regression, Ghadban Khalaf

Journal of Modern Applied Statistical Methods

During the past years, different kinds of estimators have been proposed as alternatives to the Ordinary Least Squares (OLS) estimator for the estimation of the regression coefficients in the presence of multicollinearity. In the general linear regression model, Y = Xβ + e, it is known that multicollinearity makes statistical inference difficult and may even seriously distort the inference. Ridge regression, as viewed here, defines a class of estimators of β indexed by a scalar parameter k. Two methods of specifying k are proposed and evaluated in terms of Mean Square Error (MSE) by …

Discriminating Between Generalized Exponential Distribution And Some Life Test Models Based On Population Quantiles, B. Srinivasa Rao, R. R. L Kantam

Journal of Modern Applied Statistical Methods

A test statistic based on population quantiles using sample order statistics is suggested. The quantiles of the test statistics are evaluated for generalized exponential distribution. Similar test statistic based on moments of sample order statistic is referred and the proposed test formula is compared with it. Between the pairs of the above models it is established that the test formula proposed by us is more effective and useful than the formula based on the moments of order statistics as developed by Sultan (2007).

Comparison Of Parameters Of Lognormal Distribution Based On The Classical And Posterior Estimates, Raja Sultan, S. P. Ahmad

Journal of Modern Applied Statistical Methods

Lognormal distribution is widely used in scientific field, such as agricultural, entomological, biology etc. If a variable can be thought as the multiplicative product of some positive independent random variables, then it could be modelled as lognormal. In this study, maximum likelihood estimates and posterior estimates of the parameters of lognormal distribution are obtained and using these estimates we calculate the point estimates of mean and variance for making comparisons.

On Bayesian Estimation And Predictions For Two-Component Mixture Of The Gompertz Distribution, Navid Feroze, Muhammad Aslam

Journal of Modern Applied Statistical Methods

Mixtures models have received sizeable attention from analysts in the recent years. Some work on Bayesian estimation of the parameters of mixture models have appeared. However, the were restricted to the Bayes point estimation The methodology for the Bayesian interval estimation of the parameters for said models is still to be explored. This paper proposes the posterior interval estimation (along with point estimation) for the parameters of a two-component mixture of the Gompertz distribution. The posterior predictive intervals are also derived and evaluated. Different informative and non-informative priors are assumed under a couple of loss functions for the posterior analysis. …

A Generalized Class Of Estimators For Finite Population Variance In Presence Of Measurement Errors, Prayas Sharma, Rajesh Singh

Journal of Modern Applied Statistical Methods

The problem of estimating the population variance is presented using auxiliary information in the presence of measurement errors. The estimators in this article use auxiliary information to improve efficiency and assume that measurement error is present both in study and auxiliary variable. A numerical study is carried out to compare the performance of the proposed estimator with other estimators and the variance per unit estimator in the presence of measurement errors.

Robust Regression Estimators When There Are Tied Values, Rand R. Wilcox, Florence Clark

Journal of Modern Applied Statistical Methods

It is well known that when using the ordinary least squares regression estimator, outliers among the dependent variable can result in relatively poor power. Many robust regression estimators have been derived that address this problem, but the bulk of the results assume that the dependent variable is continuous. It is demonstrated that when there are tied values, several robust regression estimators can perform poorly in terms of controlling the Type I error probability, even with a large sample size. The presence of tied values does not necessarily mean that they perform poorly, but there is the issue of whether there …