Statistics and Probability Commons^™

Confidence Intervals For Heritability Via Haseman-Elston Regression, Tamar Sofer

UW Biostatistics Working Paper Series

Heritability is the proportion of phenotypic variance in a population that is attributable to individual genotypes. Heritability is considered an important measure in both evolutionary biology and in medicine, and is routinely estimated and reported in genetic epidemiology studies. In population-based genome-wide association studies (GWAS), mixed models are used to estimate variance components, from which a heritability estimate is obtained. The estimated heritability is the proportion of the model's total variance that is due to the genetic relatedness matrix (kinship measured from genotypes). Current practice is to use bootstrapping, which is slow, or normal asymptotic approximation to estimate the precision …

Albuminuria Changes And Cardiovascular And Renal Outcomes In Type 1 Diabetes: The Dcct/Edic Study., Ian H De Boer, Xiaoyu Gao, Patricia A Cleary, Ionut Bebu, John M Lachin, Mark E Molitch, Trevor Orchard, Andrew D Paterson, Bruce A Perkins, Michael W Steffes, Bernard Zinman

Epidemiology Faculty Publications

Background and objectives In trials of people with type 2 diabetes, albuminuria reduction with renin-angiotensin system inhibitors is associated with lower risks of cardiovascular events and CKD progression. We tested whether progression or remission of microalbuminuria is associated with cardiovascular and renal risk in a well characterized cohort of type 1 diabetes.

Design, setting, participants, & measurements We studied 1441 participants in the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications study. Albumin excretion rate (AER) was quantified annually or biennially for up to 30 years. For each participant, albuminuria status was defined over time as normoalbuminuria (AER …

Generalized Least-Powers Regressions I: Bivariate Regressions, Nataniel Greene

Publications and Research

The bivariate theory of generalized least-squares is extended here to least-powers. The bivariate generalized least-powers problem of order p seeks a line which minimizes the average generalized mean of the absolute pth power deviations between the data and the line. Least-squares regressions utilize second order moments of the data to construct the regression line whereas least-powers regressions use moments of order p to construct the line. The focus is on even values of p, since this case admits analytic solution methods for the regression coefficients. A numerical example shows generalized least-powers methods performing comparably to generalized least-squares methods, …

Robust Alternatives To Ancova For Estimating The Treatment Effect Via A Randomized Comparative Study, Fei Jiang, Lu Tian, Haoda Fu, Takahiro Hasegawa, Marc Alan Pfeffer, L. J. Wei

Harvard University Biostatistics Working Paper Series

In comparing two treatments via a randomized clinical trial, the analysis of covari- ance technique is often utilized to estimate an overall treatment effect. The ANCOVA is generally perceived as a more efficient procedure than its simple two sample estima- tion counterpart. Unfortunately when the ANCOVA model is not correctly specified, the resulting estimator is generally not consistent especially when the model is nonlin- ear. Recently various nonparametric alternatives, such as the augmentation methods, to ANCOVA have been proposed to estimate the treatment effect by adjusting the covariates. However, the properties of these alternatives have not been studied in the …

Longitudinal Stability Of Effect Sizes In Education Research, Joshua Stephens

Journal of Modern Applied Statistical Methods

Educators use meta-analyses to decide best practices. It has been suggested that effect sizes have declined over time due to various biases. This study applies an established methodological framework to educational meta-analyses and finds that effect sizes have increased from 1970–present. Potential causes for this phenomenon are discussed.

The Application Of Legendre Multiwavelet Functions In Image Compression, Elham Hashemizadeh, Sohrab Rahbar

Journal of Modern Applied Statistical Methods

Legendre multiwavelets are introduced. These functions can be designed in such a way that the properties of orthogonality, polynomial approximation, and symmetry hold at the same time. In this way, they can be effectively deployed in image compression.

Fitting Flexible Parametric Regression Models With Gldreg In R, Steve Su

Journal of Modern Applied Statistical Methods

This article outlines the functionality of the GLDreg package in R which fits parametric regression models using generalized lambda distributions via maximum likelihood estimation and L moment matching. The main advantage of GLDreg is the provision of robust regression lines and smooth regression quantiles beyond the capabilities of existing known methods.

Doubly Censored Data From Two-Component Mixture Of Inverse Weibull Distributions: Theory And Applications, Tabassum Sindhu, Navid Feroze, Muhammad Aslam

Journal of Modern Applied Statistical Methods

Finite mixture distributions consist of a weighted sum of standard distributions and are a useful tool for reliability analysis of a heterogeneous population. They provide the necessary flexibility to model failure distributions of components with multiple failure models. The analysis of the mixture models under Bayesian framework has received sizable attention in the recent years. However, the Bayesian estimation of the mixture models under doubly censored samples has not yet been introduced in the literature. The main objective of this paper is to discuss the Bayes estimation of the inverse Weibull mixture distributions under doubly censoring. Different priors and loss …

Bayesian Analysis Of Discrete Skewed Laplace Distribution, A. Hossianzadeh, K Zare

Journal of Modern Applied Statistical Methods

The discrete skewed Laplace distribution is a flexible distribution with integer domain and simple closed form that can be applied to model count data. Parameters are estimated under empirical Bayes (EB) analysis and comparison are made between the Bayesian parameter estimation and classical parameter estimation, i.e. the maximum likelihood (ML) approach. The results show that the Bayesian parameter estimations are preferable.

A Comparison Of Usual T-Test Statistic And Modified T-Test Statistics On Skewed Distribution Functions, Wooi K. Lim, Alice W. Lim

Journal of Modern Applied Statistical Methods

When the sample size n is small, the random variable T= √n(\overline{X} – μ)/S is said to follow a central t distribution with degrees of freedom (n – 1), where \overline{X} is the sample mean and S is the sample standard deviation, provided that the data X ~ N (μ, σ²). The random variable T can be used as a test statistic to hypothesize the population mean μ. Some argue that the t-test statistic is robust against the normality of the distribution and claim that the normality assumption is not necessary. In this …

Designing Of Bayesian Skip Lot Sampling Plan Under Destructive Testing, K. K. Suresh, S. Umamaheswari

Journal of Modern Applied Statistical Methods

Skip-lot sampling plan serves as a cost-effective technique to manage the cost of performing frequent product inspections. As a powerful tool within a real-time quality management system, the ability to collect data which an optimize skip-lot sampling parameters affords manufacturers the luxury of lowering inspection expenses in various manufacturing units. The good quality of product can be produced in continuous improvement of production process in excellent quality history for suppliers. The procedures and necessary tables are provided for finding the respective plans for which sum of producer and consumer risks are minimized with acceptable and limiting quality levels which accounts …

Bayesian Inference For Median Of The Lognormal Distribution, K. Aruna Rao, Juliet Gratia D'Cunha

Journal of Modern Applied Statistical Methods

Lognormal distribution has many applications. The past research papers concentrated on the estimation of the mean of this distribution. This paper develops credible interval for the median of the lognormal distribution. The estimated coverage probability and average length of the credible interval is compared with the confidence interval using Monte Carlo simulation.

The Br2 – Weighting Method For Estimating The Effects Of Air Pollution On Population Health, Goran Krstic, Nikolas S. Krstic, Mauricio Zambrano-Bigiarini

Journal of Modern Applied Statistical Methods

Uncertainties, limitations and biases may impede the correct application of concentration-response linear functions to estimate the effects of air pollution exposure on population health. The reliability of a prediction depends largely on the strength of the linear correlation between the studied variables. This work proposes the joint use of the coefficient of determination, r², with the regression slope, b, as an improved measure of the strength of the linear relation between air pollution and its effects on population health. The proposed br²‑weighting method offers more reliable inferences about the potential effects of air pollution on …

Estimating The Parameter Of Exponential Distribution Under Type Ii Censoring From Fuzzy Data, Iman Makhdoom, Parviz Nasiri, Abbas Pak

Journal of Modern Applied Statistical Methods

The problem of estimating the parameter of Exponential distribution on the basis of type II censoring scheme is considered when the available data are in the form of fuzzy numbers. The Bayes estimate of the unknown parameter is obtained by using the approximation forms of Lindley (1980) and Tierney and Kadane (1986) under the assumption of gamma prior. The highest posterior density (HPD) estimate of the parameter of interest is found. A Monte Carlo simulation is used to compare the performances of the different methods. A real data set is investigated to illustrate the applicability of …

Preliminary Tests Of Normality When Comparing Three Independent Samples, Björn Lantz, Roy Andersson, Peter Manfredsson

Journal of Modern Applied Statistical Methods

This paper uses simulation to explore the performance of a two-stage procedure where a preliminary Shapiro-Wilk test is used to choose between the ANOVA and Kruskal-Wallis tests as a three-sample location test. The results suggest that the two-stage procedure actually seems to be preferable when conducting such location tests.

Limited Failure Censored Life Test Sampling Plan In Burr Type X Distribution, R. R. L. Kantam, M. S. Ravikumar

Journal of Modern Applied Statistical Methods

The Burr type X distribution is considered as a life time random variable of a product whose lots are to be decided for acceptance or otherwise on the basis of sample lifetimes drawn from the lot. The sample is divided into various groups in order to develop a group sampling plan in such a way that the life testing experiment is terminated as soon as the first failure in each group is observed. The acceptance criterion based on the theory of order statistics is proposed and is shown to be more economical than a criterion proposed in the earlier similar …

Comparison Of Some Multivariate Nonparametric Tests In Profile Analysis To Repeated Measurements, Mehrdad Vossoughi, Shila Shahvali, Erfan Sadeghi

Journal of Modern Applied Statistical Methods

Through Monte Carlo simulations, the performance of six multivariate nonparametric tests for testing the hypothesis of parallelism in profile analysis was studied. In conclusion, the tests based on ranks were as efficient as Hotelling's T² under multivariate normal distribution. For the heavy tailed distribution, the tests based on signs performed best.

On Generalizing Cumulative Ordered Regression Models, Robert W. Walker

Journal of Modern Applied Statistical Methods

We examine models that relax proportionality in cumulative ordered regression models. Something fundamental arising from ordered variables and stochastic ordering implies a partitioning. Efforts to relax proportionality also relax the ability to collapse an inherently multidimensional problem to a partitioning of the (unidimensional) real line. It is surprising and unfortunate to find that deviations from proportionality are sufficient to generate internal contradictions; undecidable propositions must exist by relaxing proportional odds without other relevant and significant changes in the underlying model. We prove a single theorem linking continuous support and partitions of a latent space to show that for these two …

Within Groups Anova When Using A Robust Multivariate Measure Of Location, Rand Wilcox, Timothy Hayes

Journal of Modern Applied Statistical Methods

For robust measures of location associated with J dependent groups, various methods have been proposed that are aimed at testing the global hypothesis of a common measure of location applied to the marginal distributions. A criticism of these methods is that they do not deal with outliers in a manner that takes into account the overall structure of the data. Location estimators have been derived that deal with outliers in this manner, but evidently there are no simulation results regarding how well they perform when the goal is to test the some global hypothesis. The paper compares four bootstrap methods …

End Matter, Jmasm Editors

Journal of Modern Applied Statistical Methods

.

Some Remarks On Rao And Lovric’S ‘Testing Point Null Hypothesis Of A Normal Mean And The Truth: 21st Century Perspective’, Bruno D. Zumbo, Edward Kroc

Journal of Modern Applied Statistical Methods

Although we have much to agree with in Rao and Lovric’s important discussion of the test of point null hypotheses, it stirred us to provide a way out of their apparent Zero probability paradox and cast the Hodges-Lehmann paradigm from a Serlin-Lapsley approach. We close our remarks with an eye toward a broad perspective.

Testing Point Null Hypothesis Of A Normal Mean And The Truth: 21st Century Perspective, Calyampudi Radhakrishna Rao, Miodrag M. Lovric

Journal of Modern Applied Statistical Methods

Testing a point (sharp) null hypothesis is arguably the most widely used statistical inferential procedure in many fields of scientific research, nevertheless, the most controversial, and misapprehended. Since 1935 when Buchanan-Wollaston raised the first criticism against hypothesis testing, this foundational field of statistics has drawn increasingly active and stronger opposition, including draconian suggestions that statistical significance testing should be abandoned or even banned. Statisticians should stop ignoring these accumulated and significant anomalies within the current point-null hypotheses paradigm and rebuild healthy foundations of statistical science. The foundation for a paradigm shift in testing statistical hypotheses is suggested, which is testing …

Study Of The Left Censored Data From The Gumbel Type Ii Distribution Under A Bayesian Approach, Tabassum Naz Sindhu, Navid Feroze, Muhammad Aslam

Journal of Modern Applied Statistical Methods

Based on left type II censored samples from a Gumbel type II distribution, the Bayes estimators and corresponding risks of the unknown parameter were obtained under different asymmetric loss functions, assuming different informative and non-informative priors. Elicitation of hyper-parameters through prior predictive approach has also been discussed. The expressions for the credible intervals and posterior predictive distributions have been derived. Comparisons of these estimators are made through simulation study using numerical and graphical methods.

Some Tests For Seasonality In Time Series Data, Eleazar Chukwunenye Nwogu, Iheanyi Sylvester Iwueze, Valentine Uchenna Nlebedim

Journal of Modern Applied Statistical Methods

This paper presents some tests for seasonality in a time series data which considers the model structure and the nature of trending curve. The tests were applied to the row variances of the Buys Ballot table. The student t-test and Wilcoxon Signed-Ranks test have been recommended for detection of seasonality.

Hierarchical Bayes Estimation Of Reliability Indexes Of Cold Standby Series System Under General Progressive Type Ii Censoring Scheme, D. R. Barot, M. N. Patel

Journal of Modern Applied Statistical Methods

In this paper, hierarchical Bayes approach is presented for estimation and prediction of reliability indexes and remaining lifetimes of a cold standby series system under general progressive Type II censoring scheme. A simulation study has been carried out for comparison purpose. The study will help reliability engineers in various industrial series system setups.

A New Estimator Of The Population Mean: An Application To Bioleaching Studies, Amer I. Al-Omari, Carlos N. Bouza, Dante Covarrubias, Roma Pal

Journal of Modern Applied Statistical Methods

The multistage balanced groups ranked set samples (MBGRSS) method is considered for estimating the population mean for samples of size m = 3k where k is a positive real integer. It is compared with the simple random sampling (SRS) and ranked set sampling (RSS) schemes. For the symmetric distributions considered in this study, the MBGRSS estimator is an unbiased estimator of the population mean and it is more efficient than SRS and RSS methods based on the same number of measured units. Its efficiency is increasing in s for fixed value of the sample size, where s is the …

A New Exponential Type Estimator For The Population Mean In Simple Random Sampling, Gamze Özel Kadilar

Journal of Modern Applied Statistical Methods

This paper provides a new exponential type estimator in simple random sampling for population mean. It is shown that proposed exponential type estimator is always more efficient than estimators considered by Bahl and Tuteja (1991) and Singh, Chauhan, Sawan, and Smarandache (2009). From numerical examples it is also observed that proposed modified ratio estimator performs better than existing estimators.

Bayesian Analysis Of Generalized Exponential Distribution, Saima Naqash, S. P. Ahmad, Aquil Ahmed

Journal of Modern Applied Statistical Methods

Bayesian estimators of unknown parameters of a two parameter generalized exponential distribution are obtained based on non-informative priors using different loss functions.

Regularized Neural Network To Identify Potential Breast Cancer: A Bayesian Approach, Hansapani S. Rodrigo, Chris P. Tsokos, Taysseer Sharaf

Journal of Modern Applied Statistical Methods

In the current study, we have exemplified the use of Bayesian neural networks for breast cancer classification using the evidence procedure. The optimal Bayesian network has 81% overall accuracy in correctly classifying the true status of breast cancer patients, 59% sensitivity in correctly detecting the malignancy and 83% specificity in correctly detecting the non-malignancy. The area under the receiver operating characteristic curve (0.7940) shows that this is a moderate classification model.

Efficient And Unbiased Estimation Procedure Of Population Mean In Two-Phase Sampling, Reba Maji, Arnab Bandyopadhyay, G. N. Singh

Journal of Modern Applied Statistical Methods

In this paper, an unbiased regression-ratio type estimator has been developed for estimating the population mean using two auxiliary variables in double sampling. Its properties are studied under two different cases. Empirical studies and graphical simulation have been done to demonstrate the efficiency of the proposed estimator over other estimators.