scholarly journals Multiple Testing. Part I. Single-Step Procedures for Control of General Type I Error Rates

2004 ◽  
Vol 3 (1) ◽  
pp. 1-69 ◽  
Author(s):  
Sandrine Dudoit ◽  
Mark J. van der Laan ◽  
Katherine S. Pollard
Keyword(s):  
Error Rate ◽  
Multiple Testing ◽  
Type I Error ◽  
Null Distribution ◽  
Error Rates ◽  
Single Step ◽  
Type I ◽  
Test Statistics ◽  
Testing Procedures ◽  
Multiple Testing Procedures

The present article proposes general single-step multiple testing procedures for controlling Type I error rates defined as arbitrary parameters of the distribution of the number of Type I errors, such as the generalized family-wise error rate. A key feature of our approach is the test statistics null distribution (rather than data generating null distribution) used to derive cut-offs (i.e., rejection regions) for these test statistics and the resulting adjusted p-values. For general null hypotheses, corresponding to submodels for the data generating distribution, we identify an asymptotic domination condition for a null distribution under which single-step common-quantile and common-cut-off procedures asymptotically control the Type I error rate, for arbitrary data generating distributions, without the need for conditions such as subset pivotality. Inspired by this general characterization of a null distribution, we then propose as an explicit null distribution the asymptotic distribution of the vector of null value shifted and scaled test statistics. In the special case of family-wise error rate (FWER) control, our method yields the single-step minP and maxT procedures, based on minima of unadjusted p-values and maxima of test statistics, respectively, with the important distinction in the choice of null distribution. Single-step procedures based on consistent estimators of the null distribution are shown to also provide asymptotic control of the Type I error rate. A general bootstrap algorithm is supplied to conveniently obtain consistent estimators of the null distribution. The special cases of t- and F-statistics are discussed in detail. The companion articles focus on step-down multiple testing procedures for control of the FWER (van der Laan et al., 2004b) and on augmentations of FWER-controlling methods to control error rates such as tail probabilities for the number of false positives and for the proportion of false positives among the rejected hypotheses (van der Laan et al., 2004a). The proposed bootstrap multiple testing procedures are evaluated by a simulation study and applied to genomic data in the fourth article of the series (Pollard et al., 2004).

scholarly journals Evaluations of the Optimal Discovery Procedure for Multiple Testing

2016 ◽  
Vol 12 (1) ◽  
pp. 21-29 ◽  
Author(s):  
Daniel B. Rubin
Keyword(s):  
Error Rate ◽  
Optimality Theory ◽  
Multiple Testing ◽  
Type I Error ◽  
Type I ◽  
Test Statistics ◽  
Testing Procedures ◽  
Type I Error Rate ◽  
Rate Measure ◽  
Optimal Discovery Procedure

Abstract The Optimal Discovery Procedure (ODP) is a method for simultaneous hypothesis testing that attempts to gain power relative to more standard techniques by exploiting multivariate structure [1]. Specializing to the example of testing whether components of a Gaussian mean vector are zero, we compare the power of the ODP to a Bonferroni-style method and to the Benjamini-Hochberg method when the testing procedures aim to respectively control certain Type I error rate measures, such as the expected number of false positives or the false discovery rate. We show through theoretical results, numerical comparisons, and two microarray examples that when the rejection regions for the ODP test statistics are chosen such that the procedure is guaranteed to uniformly control a Type I error rate measure, the technique is generally less powerful than competing methods. We contrast and explain these results in light of previously proven optimality theory for the ODP. We also compare the ordering given by the ODP test statistics to the standard rankings based on sorting univariate p-values from smallest to largest. In the cases we considered the standard ordering was superior, and ODP rankings were adversely impacted by correlation.

scholarly journals Multiple Testing. Part II. Step-Down Procedures for Control of the Family-Wise Error Rate

2004 ◽  
Vol 3 (1) ◽  
pp. 1-33 ◽  
Author(s):  
Mark J. van der Laan ◽  
Sandrine Dudoit ◽  
Katherine S. Pollard
Keyword(s):  
Error Rate ◽  
Multiple Testing ◽  
Type I Error ◽  
Null Distribution ◽  
Type I ◽  
Test Statistics ◽  
P Values ◽  
Family Wise Error Rate ◽  
The Family ◽  
Step Down

The present article proposes two step-down multiple testing procedures for asymptotic control of the family-wise error rate (FWER): the first procedure is based on maxima of test statistics (step-down maxT), while the second relies on minima of unadjusted p-values (step-down minP). A key feature of our approach is the characterization and construction of a test statistics null distribution (rather than data generating null distribution) for deriving cut-offs for these test statistics (i.e., rejection regions) and the resulting adjusted p-values. For general null hypotheses, corresponding to submodels for the data generating distribution, we identify an asymptotic domination condition for a null distribution under which the step-down maxT and minP procedures asymptotically control the Type I error rate, for arbitrary data generating distributions, without the need for conditions such as subset pivotality. Inspired by this general characterization, we then propose as an explicit null distribution the asymptotic distribution of the vector of null value shifted and scaled test statistics. Step-down procedures based on consistent estimators of the null distribution are shown to also provide asymptotic control of the Type I error rate. A general bootstrap algorithm is supplied to conveniently obtain consistent estimators of the null distribution.

scholarly journals Design, analysis and reporting of multi-arm trials and strategies to address multiple testing

2020 ◽  
Vol 49 (3) ◽  
pp. 968-978
Author(s):  
Ayodele Odutayo ◽  
Dmitry Gryaznov ◽  
Bethan Copsey ◽  
Paul Monk ◽  
Benjamin Speich ◽  
...  
Keyword(s):  
Multiple Testing ◽  
Type I Error ◽  
Testing Procedure ◽  
Type I ◽  
Decision Tool ◽  
Multiple Testing Procedure ◽  
Testing Procedures ◽  
Global Comparison ◽  
Multiple Testing Procedures ◽  
Trial Protocols

Abstract Background It is unclear how multiple treatment comparisons are managed in the analysis of multi-arm trials, particularly related to reducing type I (false positive) and type II errors (false negative). Methods We conducted a cohort study of clinical-trial protocols that were approved by research ethics committees in the UK, Switzerland, Germany and Canada in 2012. We examined the use of multiple-testing procedures to control the overall type I error rate. We created a decision tool to determine the need for multiple-testing procedures. We compared the result of the decision tool to the analysis plan in the protocol. We also compared the pre-specified analysis plans in trial protocols to their publications. Results Sixty-four protocols for multi-arm trials were identified, of which 50 involved multiple testing. Nine of 50 trials (18%) used a single-step multiple-testing procedures such as a Bonferroni correction and 17 (38%) used an ordered sequence of primary comparisons to control the overall type I error. Based on our decision tool, 45 of 50 protocols (90%) required use of a multiple-testing procedure but only 28 of the 45 (62%) accounted for multiplicity in their analysis or provided a rationale if no multiple-testing procedure was used. We identified 32 protocol–publication pairs, of which 8 planned a global-comparison test and 20 planned a multiple-testing procedure in their trial protocol. However, four of these eight trials (50%) did not use the global-comparison test. Likewise, 3 of the 20 trials (15%) did not perform the multiple-testing procedure in the publication. The sample size of our study was small and we did not have access to statistical-analysis plans for the included trials in our study. Conclusions Strategies to reduce type I and type II errors are inconsistently employed in multi-arm trials. Important analytical differences exist between planned analyses in clinical-trial protocols and subsequent publications, which may suggest selective reporting of analyses.

scholarly journals Adding new experimental arms to randomised clinical trials: Impact on error rates

2020 ◽  
Vol 17 (3) ◽  
pp. 273-284 ◽  
Author(s):  
Babak Choodari-Oskooei ◽  
Daniel J Bratton ◽  
Melissa R Gannon ◽  
Angela M Meade ◽  
Matthew R Sydes ◽  
...  
Keyword(s):  
Error Rate ◽  
Type I Error ◽  
Controlled Trial ◽  
Late Phase ◽  
Pairwise Comparison ◽  
Error Rates ◽  
Pairwise Comparisons ◽  
Type I ◽  
Test Statistics ◽  
Type I Error Rate

Background: Experimental treatments pass through various stages of development. If a treatment passes through early-phase experiments, the investigators may want to assess it in a late-phase randomised controlled trial. An efficient way to do this is adding it as a new research arm to an ongoing trial while the existing research arms continue, a so-called multi-arm platform trial. The familywise type I error rate is often a key quantity of interest in any multi-arm platform trial. We set out to clarify how it should be calculated when new arms are added to a trial some time after it has started. Methods: We show how the familywise type I error rate, any-pair and all-pairs powers can be calculated when a new arm is added to a platform trial. We extend the Dunnett probability and derive analytical formulae for the correlation between the test statistics of the existing pairwise comparison and that of the newly added arm. We also verify our analytical derivation via simulations. Results: Our results indicate that the familywise type I error rate depends on the shared control arm information (i.e. individuals in continuous and binary outcomes and primary outcome events in time-to-event outcomes) from the common control arm patients and the allocation ratio. The familywise type I error rate is driven more by the number of pairwise comparisons and the corresponding (pairwise) type I error rates than by the timing of the addition of the new arms. The familywise type I error rate can be estimated using Šidák’s correction if the correlation between the test statistics of pairwise comparisons is less than 0.30. Conclusions: The findings we present in this article can be used to design trials with pre-planned deferred arms or to add new pairwise comparisons within an ongoing platform trial where control of the pairwise error rate or familywise type I error rate (for a subset of pairwise comparisons) is required.

Correction: “Influence of Selection Bias on the Test Decision – A Simulation Study”

2014 ◽  
Vol 53 (05) ◽  
pp. 343-343
Keyword(s):  
Selection Bias ◽  
Simulation Study ◽  
Error Rate ◽  
Type I Error ◽  
Block Size ◽  
Error Rates ◽  
Type I ◽  
Type I Error Rate ◽  
Representation Error ◽  
Numeric Representation

We have to report marginal changes in the empirical type I error rates for the cut-offs 2/3 and 4/7 of Table 4, Table 5 and Table 6 of the paper “Influence of Selection Bias on the Test Decision – A Simulation Study” by M. Tamm, E. Cramer, L. N. Kennes, N. Heussen (Methods Inf Med 2012; 51: 138 –143). In a small number of cases the kind of representation of numeric values in SAS has resulted in wrong categorization due to a numeric representation error of differences. We corrected the simulation by using the round function of SAS in the calculation process with the same seeds as before. For Table 4 the value for the cut-off 2/3 changes from 0.180323 to 0.153494. For Table 5 the value for the cut-off 4/7 changes from 0.144729 to 0.139626 and the value for the cut-off 2/3 changes from 0.114885 to 0.101773. For Table 6 the value for the cut-off 4/7 changes from 0.125528 to 0.122144 and the value for the cut-off 2/3 changes from 0.099488 to 0.090828. The sentence on p. 141 “E.g. for block size 4 and q = 2/3 the type I error rate is 18% (Table 4).” has to be replaced by “E.g. for block size 4 and q = 2/3 the type I error rate is 15.3% (Table 4).”. There were only minor changes smaller than 0.03. These changes do not affect the interpretation of the results or our recommendations.

scholarly journals The Romano–Wolf multiple-hypothesis correction in Stata

2020 ◽  
Vol 20 (4) ◽  
pp. 812-843
Author(s):  
Damian Clarke ◽  
Joseph P. Romano ◽  
Michael Wolf
Keyword(s):  
Error Rate ◽  
Multiple Testing ◽  
Original Data ◽  
Dependence Structure ◽  
Familywise Error Rate ◽  
Testing Procedures ◽  
Multiple Testing Procedures ◽  
Multiple Hypothesis ◽  
Wide Range ◽  
Performance Gains

When considering multiple-hypothesis tests simultaneously, standard statistical techniques will lead to overrejection of null hypotheses unless the multiplicity of the testing framework is explicitly considered. In this article, we discuss the Romano–Wolf multiple-hypothesis correction and document its implementation in Stata. The Romano–Wolf correction (asymptotically) controls the familywise error rate, that is, the probability of rejecting at least one true null hypothesis among a family of hypotheses under test. This correction is considerably more powerful than earlier multiple-testing procedures, such as the Bonferroni and Holm corrections, given that it takes into account the dependence structure of the test statistics by resampling from the original data. We describe a command, rwolf, that implements this correction and provide several examples based on a wide range of models. We document and discuss the performance gains from using rwolf over other multiple-testing procedures that control the familywise error rate.

scholarly journals Efficient Noninferiority Testing Procedures for Simultaneously Assessing Sensitivity and Specificity of Two Diagnostic Tests

2015 ◽  
Vol 2015 ◽  
pp. 1-7 ◽  
Author(s):  
Guogen Shan ◽  
Amei Amei ◽  
Daniel Young
Keyword(s):  
Sensitivity And Specificity ◽  
Diagnostic Tests ◽  
Error Control ◽  
Type I Error ◽  
Binary Outcomes ◽  
Type I ◽  
Test Statistics ◽  
Asymptotic Approach ◽  
Testing Procedures ◽  
Assessing Sensitivity

Sensitivity and specificity are often used to assess the performance of a diagnostic test with binary outcomes. Wald-type test statistics have been proposed for testing sensitivity and specificity individually. In the presence of a gold standard, simultaneous comparison between two diagnostic tests for noninferiority of sensitivity and specificity based on an asymptotic approach has been studied by Chen et al. (2003). However, the asymptotic approach may suffer from unsatisfactory type I error control as observed from many studies, especially in small to medium sample settings. In this paper, we compare three unconditional approaches for simultaneously testing sensitivity and specificity. They are approaches based on estimation, maximization, and a combination of estimation and maximization. Although the estimation approach does not guarantee type I error, it has satisfactory performance with regard to type I error control. The other two unconditional approaches are exact. The approach based on estimation and maximization is generally more powerful than the approach based on maximization.

Comparison of methods to account for autocorrelation in correlation analyses of fish data

1998 ◽  
Vol 55 (9) ◽  
pp. 2127-2140 ◽  
Author(s):  
Brian J Pyper ◽  
Randall M Peterman
Keyword(s):  
Monte Carlo ◽  
Hypothesis Testing ◽  
Type I Error ◽  
Low Frequency ◽  
Error Rates ◽  
Type I ◽  
Testing Procedures ◽  
Type I Error Rates ◽  
Fish Recruitment ◽  
Correlation Analyses

Autocorrelation in fish recruitment and environmental data can complicate statistical inference in correlation analyses. To address this problem, researchers often either adjust hypothesis testing procedures (e.g., adjust degrees of freedom) to account for autocorrelation or remove the autocorrelation using prewhitening or first-differencing before analysis. However, the effectiveness of methods that adjust hypothesis testing procedures has not yet been fully explored quantitatively. We therefore compared several adjustment methods via Monte Carlo simulation and found that a modified version of these methods kept Type I error rates near . In contrast, methods that remove autocorrelation control Type I error rates well but may in some circumstances increase Type II error rates (probability of failing to detect some environmental effect) and hence reduce statistical power, in comparison with adjusting the test procedure. Specifically, our Monte Carlo simulations show that prewhitening and especially first-differencing decrease power in the common situations where low-frequency (slowly changing) processes are important sources of covariation in fish recruitment or in environmental variables. Conversely, removing autocorrelation can increase power when low-frequency processes account for only some of the covariation. We therefore recommend that researchers carefully consider the importance of different time scales of variability when analyzing autocorrelated data.

Sign in / Sign up

Export Citation Format

Share Document