Familywise Robustness Criteria Revisited for Newer Multiple Testing Procedures PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Familywise Robustness Criteria Revisited for Newer Multiple Testing Procedures PDF full book. Access full book title Familywise Robustness Criteria Revisited for Newer Multiple Testing Procedures by Charles W. Miller. Download full books in PDF and EPUB format.
Author: Charles W. Miller Publisher: ISBN: Category : Languages : en Pages : 101
Book Description
As the availability of large datasets becomes more prevalent, so does the need to discover significant findings among a large collection of hypotheses. Multiple testing procedures (MTP) are used to control the familywise error rate (FWER) or the chance to commit at least one type I error when performing multiple hypotheses testing. When controlling the FWER, the power of a MTP to detect significant differences decreases as the number of hypotheses increases. It would be ideal to discover the same false null hypotheses despite the family of hypotheses chosen to be tested. Holland and Cheung (2002) developed measures called familywise robustness criteria (FWR) to study the effect of family size on the acceptance and rejection of a hypothesis. Their analysis focused on procedures that controlled FWER and false discovery rate (FDR). Newer MTPs have since been developed which control the generalized FWER (gFWER (k) or k-FWER) and false discovery proportion (FDP) or tail probabilities for the proportion of false positives (TPPFP). This dissertation reviews these newer procedures and then discusses the effect of family size using the FWRs of Holland and Cheung. In the case where the test statistics are independent and the null hypotheses are all true, the Type R enlargement familywise robustness measure can be expressed as a ratio of the expected number of Type I errors. In simulations, positive dependence among the test statistics was introduced, the expected number of Type I errors and the Type R enlargement FWR increased for step-up procedures with higher levels of correlation, but not for step-down or single-step procedures.
Author: Charles W. Miller Publisher: ISBN: Category : Languages : en Pages : 101
Book Description
As the availability of large datasets becomes more prevalent, so does the need to discover significant findings among a large collection of hypotheses. Multiple testing procedures (MTP) are used to control the familywise error rate (FWER) or the chance to commit at least one type I error when performing multiple hypotheses testing. When controlling the FWER, the power of a MTP to detect significant differences decreases as the number of hypotheses increases. It would be ideal to discover the same false null hypotheses despite the family of hypotheses chosen to be tested. Holland and Cheung (2002) developed measures called familywise robustness criteria (FWR) to study the effect of family size on the acceptance and rejection of a hypothesis. Their analysis focused on procedures that controlled FWER and false discovery rate (FDR). Newer MTPs have since been developed which control the generalized FWER (gFWER (k) or k-FWER) and false discovery proportion (FDP) or tail probabilities for the proportion of false positives (TPPFP). This dissertation reviews these newer procedures and then discusses the effect of family size using the FWRs of Holland and Cheung. In the case where the test statistics are independent and the null hypotheses are all true, the Type R enlargement familywise robustness measure can be expressed as a ratio of the expected number of Type I errors. In simulations, positive dependence among the test statistics was introduced, the expected number of Type I errors and the Type R enlargement FWR increased for step-up procedures with higher levels of correlation, but not for step-down or single-step procedures.
Author: Zijiang Yang Publisher: ISBN: Category : Languages : en Pages : 91
Book Description
The main research topic in this dissertation is the development of the closure method of multiple testing procedures. Considering a general procedure that allows the underlying test statistics as well as the associated parameters to be dependent, we first propose a step-down procedure controlling the FWER, which is defined as the probability of committing at least one false discovery. Holm (1979) first proposed a step-down procedure for multiple hypothesis testing with a control of the familywise error rate (FWER) under any kind of dependence. Under the normal distributional setup, Seneta and Chen (2005) sharpened the Holm procedure by taking into account the correlations between the test statistics. In this dissertation, the Seneta-Chen procedure is further modified yielding a more powerful FWER controlling procedure. We then advance our research and propose another step-down procedure to control the generalized FWER (k-FWER), which is defined as the probability of making at least k false discoveries. We compare our proposed k-FWER procedure with the Lehmann and Romano (2005) procedure. The proposed k-FWER procedure is more powerful, particularly when there is a strong dependence in the tests. When the proportion of true null hypotheses is expected to be small, the traditional tests are usually conservative by a factor associated with pi0, which is the proportion of true null hypotheses among all null hypotheses. Under independence, two procedures controlling the FWER and the k-FWER are proposed in this dissertation. Simulations are carried out to show that our procedures often provide much better FWER or k-FWER control and power than the traditional procedures.
Author: Yiyong Fu Publisher: ISBN: Category : Languages : en Pages : 122
Book Description
The importance of multiplicity adjustment has gained wide recognition in modern scientific research. Without it, there will be too many spurious results and reproducibility becomes an issue; with it, if overtly conservative, discoveries will be made more difficult. In the current literature on repeated testing of multiple hypotheses, Bonferroni-based methods are still the main vehicle carrying the bulk of multiplicity adjustment. There is room for power improvement by suitably utilizing both hypothesis-wise and analysis- wise dependencies. This research will contribute to the development of a natural group-sequential extension of the classical stepwise multiple testing procedures, such as Dunnett's stepdown and Hochberg's step-up procedures. It is shown that the proposed group-sequential procedures strongly control the familywise error rate while being more powerful than the recently developed class of group-sequential Bonferroni-Holm's procedures. Particularly in this research, a convexity property is discovered for the distribution of the maxima of pairwise null P-values with the underlying test statistics having distributions such as bivariate normal, t, Gamma, F, or Archimedean copulas. Such property renders itself for an immediate use in improving Holm's procedure by incorporating pairwise dependencies of P-values. The improved Holm's procedure, as all stepdown multiple testing procedures, can also be naturally extended to group-sequential setting.
Author: Alex Dmitrienko Publisher: CRC Press ISBN: 1584889853 Category : Mathematics Languages : en Pages : 323
Book Description
Useful Statistical Approaches for Addressing Multiplicity IssuesIncludes practical examples from recent trials Bringing together leading statisticians, scientists, and clinicians from the pharmaceutical industry, academia, and regulatory agencies, Multiple Testing Problems in Pharmaceutical Statistics explores the rapidly growing area of multiple c
Author: Lilun Du Publisher: ISBN: Category : Languages : en Pages : 134
Book Description
In the context of large-scale multiple testing, hypotheses are often accompanied with certain prior information. In chapter 2, we present a single-index modulated multiple testing procedure, which maintains control of the false discovery rate while incorporating prior information, by assuming the availability of a bivariate p-value for each hypothesis. To find the optimal rejection region for the bivariate p-value, we propose a criteria based on the ratio of probability density functions of the bivariate p-value under the true null and non-null. This criteria in the bivariate normal setting further motivates us to project the bivariate p-value to a single index p-value, for a wide range of directions. The true null distribution of the single index p-value is estimated via parametric and nonparametric approaches, leading to two procedures for estimating and controlling the false discovery rate. To derive the optimal projection direction, we propose a new approach based on power comparison, which is further shown to be consistent under some mild conditions. Multiple testing based on chi-squared test statistics is commonly used in many scientific fields such as genomics research and brain imaging studies. However, the challenges associated with designing a formal testing procedure when there exists a general dependence structure across the chi-squared test statistics have not been well addressed. In chapter 3, we propose a Factor Connected procedure to fill in this gap. We first adopt a latent factor structure to construct a testing framework for approximating the false discovery proportion (FDP) for a large number of highly correlated chi-squared test statistics with finite degrees of freedom k. The testing framework is then connected to simultaneously testing k linear constraints in a large dimensional linear factor model involved with some observable and unobservable common factors, resulting in a consistent estimator of FDP based on the associated unadjusted p-values.
Author: Frank Bretz Publisher: CRC Press ISBN: 1420010905 Category : Mathematics Languages : en Pages : 202
Book Description
Adopting a unifying theme based on maximum statistics, Multiple Comparisons Using R describes the common underlying theory of multiple comparison procedures through numerous examples. It also presents a detailed description of available software implementations in R. The R packages and source code for the analyses are available at http://CRAN.R-project.org After giving examples of multiplicity problems, the book covers general concepts and basic multiple comparisons procedures, including the Bonferroni method and Simes’ test. It then shows how to perform parametric multiple comparisons in standard linear models and general parametric models. It also introduces the multcomp package in R, which offers a convenient interface to perform multiple comparisons in a general context. Following this theoretical framework, the book explores applications involving the Dunnett test, Tukey’s all pairwise comparisons, and general multiple contrast tests for standard regression models, mixed-effects models, and parametric survival models. The last chapter reviews other multiple comparison procedures, such as resampling-based procedures, methods for group sequential or adaptive designs, and the combination of multiple comparison procedures with modeling techniques. Controlling multiplicity in experiments ensures better decision making and safeguards against false claims. A self-contained introduction to multiple comparison procedures, this book offers strategies for constructing the procedures and illustrates the framework for multiple hypotheses testing in general parametric models. It is suitable for readers with R experience but limited knowledge of multiple comparison procedures and vice versa. See Dr. Bretz discuss the book.
Author: Rand Wilcox Publisher: CRC Press ISBN: 149879680X Category : Mathematics Languages : en Pages : 947
Book Description
Requiring no prior training, Modern Statistics for the Social and Behavioral Sciences provides a two-semester, graduate-level introduction to basic statistical techniques that takes into account recent advances and insights that are typically ignored in an introductory course. Hundreds of journal articles make it clear that basic techniques, routinely taught and used, can perform poorly when dealing with skewed distributions, outliers, heteroscedasticity (unequal variances) and curvature. Methods for dealing with these concerns have been derived and can provide a deeper, more accurate and more nuanced understanding of data. A conceptual basis is provided for understanding when and why standard methods can have poor power and yield misleading measures of effect size. Modern techniques for dealing with known concerns are described and illustrated. Features: Presents an in-depth description of both classic and modern methods Explains and illustrates why recent advances can provide more power and a deeper understanding of data Provides numerous illustrations using the software R Includes an R package with over 1300 functions Includes a solution manual giving detailed answers to all of the exercises This second edition describes many recent advances relevant to basic techniques. For example, a vast array of new and improved methods is now available for dealing with regression, including substantially improved ANCOVA techniques. The coverage of multiple comparison procedures has been expanded and new ANOVA techniques are described. Rand Wilcox is a professor of psychology at the University of Southern California. He is the author of 13 other statistics books and the creator of the R package WRS. He currently serves as an associate editor for five statistics journals. He is a fellow of the Association for Psychological Science and an elected member of the International Statistical Institute.
Author: Daniel Navarro Publisher: Lulu.com ISBN: 1326189727 Category : Computers Languages : en Pages : 617
Book Description
"Learning Statistics with R" covers the contents of an introductory statistics class, as typically taught to undergraduate psychology students, focusing on the use of the R statistical software and adopting a light, conversational style throughout. The book discusses how to get started in R, and gives an introduction to data manipulation and writing scripts. From a statistical perspective, the book discusses descriptive statistics and graphing first, followed by chapters on probability theory, sampling and estimation, and null hypothesis testing. After introducing the theory, the book covers the analysis of contingency tables, t-tests, ANOVAs and regression. Bayesian statistics are covered at the end of the book. For more information (and the opportunity to check the book out before you buy!) visit http://ua.edu.au/ccs/teaching/lsr or http://learningstatisticswithr.com
Author: Charles W. Miller
Author: Charles W. Miller
Author: Ka Man Chan
Author: Zijiang Yang
Author: Elliot Williams
Author: Yiyong Fu
Author: Alex Dmitrienko
Author: Lilun Du
Author: Frank Bretz
Author: Rand Wilcox
Author: Daniel Navarro