A Modified MANN-WHITNEY Test with Improved Asymptotic Relative Efficiency

A slight modification of the MANN -WHITNEY test is introduced here. When the underlying distribution is normal the ARE of the modified test with respect to t-test is 99.22% (MANN -WHITNEY test: 95.45%). The lower bound of the ARE of the modified test is 92.50% for all distributions with a continuous density function and a finite variance (MANN -WHITNEY test: 86.4%).     

Testing approaches for overdispersion in poisson regression versus the generalized poisson model

Overdispersion is a common phenomenon in Poisson modeling, and the negative binomial (NB) model is frequently used to account for overdispersion. Testing approaches (Wald test, likelihood ratio test (LRT), and score test) for overdispersion in the Poisson regression versus the NB model are available. Because the generalized Poisson (GP) model is similar to the NB model, we consider the former as an alternate model for overdispersed count data. The score test has an advantage over the LRT and the Wald test in that the score test only requires that the parameter of interest be estimated under the null hypothesis. This paper proposes a score test for overdispersion based on the GP model and compares the power of the test with the LRT and Wald tests. A simulation study indicates the score test based on asymptotic standard Normal distribution is more appropriate in practical application for higher empirical power, however, it underestimates the nominal significance level, especially in small sample situations, and examples illustrate the results of comparing the candidate tests between the Poisson and GP models. A bootstrap test is also proposed to adjust the underestimation of nominal level in the score statistic when the sample size is small. The simulation study indicates the bootstrap test has significance level closer to nominal size and has uniformly greater power than the score test based on asymptotic standard Normal distribution. From a practical perspective, we suggest that, if the score test gives even a weak indication that the Poisson model is inappropriate, say at the 0.10 significance level, we advise the more accurate bootstrap procedure as a better test for comparing whether the GP model is more appropriate than Poisson model. Finally, the Vuong test is illustrated to choose between GP and NB2 models for the same dataset.     

Fluorescent virus precipitin test.

A fluorescent virus precipitin test (FVPT) for the serologic identification of small particulate antigens such as viruses has been described. The test has several advantageous characteristics: (a) It is probably as sensitive as any serologic test (i.e., aggregates with dimensions of 0.2 mum are detectable; therefore, complexes containing as few as three large viruses would give a positive test). (b) Cultivation of the virus is not required. (c) Since an indirect test can be used, only a single fluorescent conjugate is needed to permit the detection of a number of viruses. (d) The indirect test can be used to detect antiviral antibody. (e) The FVPT is rapid and reliable. (f) Its simplicity should enhance its general acceptance and application.     

A note on the tests for clustered matched-pair binary data

McNemar's test is used to assess the difference between two different procedures (treatments) using independent matched-pair data. For matched-pair data collected in clusters, the tests proposed by Durkalski et al. and Obuchowski are popular and commonly used in practice since these tests do not require distributional assumptions or assumptions on the structure of the within-cluster correlation of the data. Motivated by these tests, this note proposes a modified Obuchowski test and illustrates comparisons of the proposed test with the extant methods. An extensive Monte Carlo simulation study suggests that the proposed test performs well with respect to the nominal size, and has higher power; Obuchowski's test is most conservative, and the performance of the Durkalski's test varies between the modified Obuchowski test and the original Obuchowski's test. These results form the basis for our recommendation that (i) for equal cluster size, the modified Obuchowski test is always preferred; (ii) for varying cluster size Durkalski's test can be used for a small number of clusters (e.g. K < 50), whereas for a large number of clusters (e.g. K ≥ 50) the modified Obuchowski test is preferred. Finally, to illustrate practical application of the competing tests, two real collections of clustered matched-pair data are analyzed.     

Test retest reliability and minimal detectable change of a novel submaximal graded exercise test in the measurement of graded exercise test duration

Measurement of graded exercise test duration is clinically important and can be assessed by maximal graded exercise testing. Yet, limitations of maximal graded exercise testing exist. An alternative to maximal graded exercise testing is submaximal graded exercise testing. However, no studies have investigated the reliability of a submaximal graded exercise test in the measurement of graded exercise test duration. The purpose of this study was to determine the test-retest reliability and minimal detectable change (MDC) of a novel submaximal graded exercise test in the measurement of graded exercise test duration. Fifteen people (4 men, 11 women) with a mean age of 26.20 years (SD = 9.04) participated in this study. A novel submaximal graded exercise test was used to measure graded exercise test duration for each participant. Endpoints of the test were either 85% of age-predicted maximum heart rate or voluntarily stopping the test, whichever endpoint occurred first. Heart rate and graded exercise test duration were constantly measured throughout the test. Graded exercise test duration was defined as the total duration (minutes) of the test. For all participants, the submaximal graded exercise test was conducted at baseline and 48-72 hours thereafter. The intraclass correlation coefficient for the test-retest reliability of the test in determining graded exercise test duration was 0.94 (95% CI = 0.83-0.98). The MDC of the test in the measurement of graded exercise test duration was 0.86 minutes. The results suggest that clinicians can use this novel submaximal graded exercise test to reliably measure graded exercise test duration with a measurement error, as expressed by the MDC, of 0.86 minutes.     

A Modified Exact Test for 2 × 2 Contingency Tables

A modified exact test is proposed for 2×2 contingency tables. This test, which is based on a less conservative definition of the concept of significance (STONE, 1969) is compared with a modified form of Pearson's X² test and with Tocher's randomized exact (UMPU) test. The sizes of the new test lie near the nominal 0.05 levels while those of the X² test usually exceed the nominal level, sometimes by a factor of 2 or more. The power of the modified test is usually close to that of the UMPU test.     

A Comparison between the LEHMACHER & WALL Rank Tests and Pyhel's Permutation Test for the Analysis of r Independent Samples of Response Curves

LEHMACHER & WALL'S (1978) example of the application of a rank test for the comparison of two independent samples of response curves is reanalyzed by PYHEL'S (1980) permutation test for the hypothesis of parallelism of response curves. This permutation test is part of a complete evaluation of effects for a split-plot design using the permutation test based procedure by WILLMES & PYHEL (1981). Differences in test decisions are discussed.     

On testing departure from the binomial and multinomial assumptions

This paper is concerned with testing the multinomial (binomial) assumption against the Dirichlet-multinomial (beta-binomial) alternatives. In particular, we discuss the distribution of the asymptotic likelihood ratio (LR) test and obtain the C(alpha) goodness-of-fit test statistic. The inadequacy of the regular chi-square approximation to the LR test is supported by some Monte Carlo experiments. The C(alpha) test is recommended based on empirical significance level and power and also computational simplicity. Two examples are given.     

A goodness-of-fit test for capture-recapture model Mt under closure

A new, fully efficient goodness-of-fit test for the time-specific closed-population capture-recapture model Mt is presented. This test is based on the residual distribution of the capture history data given the maximum likelihood parameter estimates under model Mt, is partitioned into informative components, and is based on chi-square statistics. Comparison of this test with Leslie's test (Leslie, 1958, Journal of Animal Ecology 27, 84-86) for model Mt, using Monte Carlo simulations, shows the new test generally outperforms Leslie's test. The new test is frequently computable when Leslie's test is not, has Type I error rates that are closer to nominal error rates than Leslie's test, and is sensitive to behavioral variation and heterogeneity in capture probabilities. Leslie's test is not sensitive to behavioral variation in capture probabilities but, when computable, has greater power to detect heterogeneity than the new test.     

A New Rank Test Against Ordered Alternatives in K-Sample Problems

The distribution-free test against ordered alternatives proposed by Jonckheere (1954) is based on the Kendall's rank correlation coefficient τ. A new rank test is proposed and illustrated with a numerical example. The proposed test is based on Spearman's σ and has similar functional structure as the Kruskal-Wallis test. A useful by-product is a test for departure from a trend.     

Linkage detection adaptive to linkage disequilibrium: the disequilibrium maximum-likelihood-binomial test for affected-sibship data

It has been demonstrated in the literature that the transmission/disequilibrium test (TDT) has higher power than the affected-sib-pair (ASP) mean test when linkage disequilibrium (LD) is strong but that the mean test has higher power when LD is weak. Thus, for ASP data, it seems clear that the TDT should be used when LD is strong but that the mean test or other linkage tests should be used when LD is weak or absent. However, in practice, it may be difficult to follow such a guideline, because the extent of LD is often unknown. Even with a highly dense genetic-marker map, in which some markers should be located near the disease-predisposing mutation, strong LD is not inevitable. Besides the genetic distance, LD is also affected by many factors, such as the allelic heterogeneity at the disease locus, the initial LD, the allelic frequencies at both disease locus and marker locus, and the age of the mutation. Therefore, it is of interest to develop methods that are adaptive to the extent of LD. In this report, we propose a disequilibrium maximum-binomial-likelihood (DMLB) test that incorporates LD in the maximum-binomial-likelihood (MLB) test. Examination of the corresponding score statistics shows that this method adaptively combines two sources of information: (a) the identity-by-descent (IBD) sharing score, which is informative for linkage regardless of the existence of LD, and (b) the contrast between allele-specific IBD sharing score, which is informative for linkage only in the presence of LD. For ASP data, the proposed test has higher power than either the TDT or the mean test when the extent of LD ranges from moderate to strong. Only when LD is very weak or absent is the DMLB slightly less powerful than the mean test; in such cases, the TDT has essentially no power to detect linkage. Therefore, the DMLB test is an interesting approach to linkage detection when the extent of LD is unknown.     

An approximately unbiased test of phylogenetic tree selection

An approximately unbiased (AU) test that uses a newly devised multiscale bootstrap technique was developed for general hypothesis testing of regions in an attempt to reduce test bias. It was applied to maximum-likelihood tree selection for obtaining the confidence set of trees. The AU test is based on the theory of Efron et al. (Proc. Natl. Acad. Sci. USA 93:13429-13434; 1996), but the new method provides higher-order accuracy yet simpler implementation. The AU test, like the Shimodaira-Hasegawa (SH) test, adjusts the selection bias overlooked in the standard use of the bootstrap probability and Kishino-Hasegawa tests. The selection bias comes from comparing many trees at the same time and often leads to overconfidence in the wrong trees. The SH test, though safe to use, may exhibit another type of bias such that it appears conservative. Here I show that the AU test is less biased than other methods in typical cases of tree selection. These points are illustrated in a simulation study as well as in the analysis of mammalian mitochondrial protein sequences. The theoretical argument provides a simple formula that covers the bootstrap probability test, the Kishino-Hasegawa test, the AU test, and the Zharkikh-Li test. A practical suggestion is provided as to which test should be used under particular circumstances.     

Superiority inferences on individual endpoints following noninferiority testing in clinical trials

We consider the problem of drawing superiority inferences on individual endpoints following non-inferiority testing. R?hmel et al. (2006) pointed out this as an important problem which had not been addressed by the previous procedures that only tested for global superiority. R?hmel et al. objected to incorporating the non-inferiority tests in the assessment of the global superiority test by exploiting the relationship between the two, since the results of the latter test then depend on the non-inferiority margins specified for the former test. We argue that this is justified, besides the fact that it enhances the power of the global superiority test. We provide a closed testing formulation which generalizes the three-step procedure proposed by R?hmel et al. for two endpoints. For the global superiority test, R?hmel et al. suggest using the L?uter (1996) test which is modified to make it monotone. The resulting test not only is complicated to use, but the modification does not readily extend to more than two endpoints, and it is less powerful in general than several of its competitors. This is verified in a simulation study. Instead, we suggest applying the one-sided likelihood ratio test used by Perlman and Wu (2004) or the union-intersection t(max) test used by Tamhane and Logan (2004).     

A New Nonparametric Approach to the Comparison of K Independent Samples of Response Curves II: A K Sample Generalization of the FRIEDMAN Test

A K sample generalization of the FRIEDMAN test (1937) is introduced which can be used as a nonparametric procedure for testing the homogeneity of the profiles of K independent samples of response curves measured at T identical points of time. While a similar procedure in LEHMACHER & WALL (1978), section 3, is based on T combined tests, each of them at level a/T, here a finite and asymptotic test is presented which is based on a single test statistic. The application of the new multivariate test is illustrated by the same numerical example as in LEHMACHER & WALL (1978). The properties of this test are discussed and compared with the combined test mentioned above.     

A modified sign test for comparing paired ROC curves

We develop a permutation test for assessing a difference in the areas under the curve (AUCs) in a paired setting where both modalities are given to each diseased and nondiseased subject. We propose that permutations be made between subjects specifically by shuffling the diseased/nondiseased labels of the subjects within each modality. As these permutations are made within modality, the permutation test is valid even if both modalities are measured on different scales. We show that our permutation test is a sign test for the symmetry of an underlying discrete distribution whose size remains valid under the assumption of equal AUCs. We demonstrate the operating characteristics of our test via simulation and show that our test is equal in power to a permutation test recently proposed by Bandos and others (2005).     

Trimmed Weighted Simes' Test for Two One‐Sided Hypotheses With Arbitrarily Correlated Test Statistics

The two‐sided Simes test is known to control the type I error rate with bivariate normal test statistics. For one‐sided hypotheses, control of the type I error rate requires that the correlation between the bivariate normal test statistics is non‐negative. In this article, we introduce a trimmed version of the one‐sided weighted Simes test for two hypotheses which rejects if (i) the one‐sided weighted Simes test rejects and (ii) both p‐values are below one minus the respective weighted Bonferroni adjusted level. We show that the trimmed version controls the type I error rate at nominal significance level α if (i) the common distribution of test statistics is point symmetric and (ii) the two‐sided weighted Simes test at level 2α controls the level. These assumptions apply, for instance, to bivariate normal test statistics with arbitrary correlation. In a simulation study, we compare the power of the trimmed weighted Simes test with the power of the weighted Bonferroni test and the untrimmed weighted Simes test. An additional result of this article ensures type I error rate control of the usual weighted Simes test under a weak version of the positive regression dependence condition for the case of two hypotheses. This condition is shown to apply to the two‐sided p‐values of one‐ or two‐sample t‐tests for bivariate normal endpoints with arbitrary correlation and to the corresponding one‐sided p‐values if the correlation is non‐negative. The Simes test for such types of bivariate t‐tests has not been considered before. According to our main result, the trimmed version of the weighted Simes test then also applies to the one‐sided bivariate t‐test with arbitrary correlation.     

Testing substitution models within a phylogenetic tree

Phylogenetic tree reconstruction frequently assumes the homogeneity of the substitution process over the whole tree. To test this assumption statistically, we propose a test based on the sample covariance matrix of the set of substitution rate matrices estimated from pairwise sequence comparison. The sample covariance matrix is condensed into a one-dimensional test statistic Delta = sum ln(1 + delta(i)), where delta(i) are the eigenvalues of the sample covariance matrix. The test does not assume a specific mutational model. It analyses the variation in the estimated rate matrices. The distribution of this test statistic is determined by simulations based on the phylogeny estimated from the data. We study the power of the test under various scenarios and apply the test to X chromosome and mtDNA primate sequence data. Finally, we demonstrate how to include rate variation in the test.     

A simple numerical approach towards improving the two one-sided test for average bioequivalence

For testing average bioequivalence, a simple test that improves upon the two-one sided test (TOST) is derived. The new test uses a critical value that is a non-increasing continuous function of the sample standard deviation, and the function is numerically obtained. Simulation results show that the resulting test performs well in terms of type I error probability; in particular, the test avoids the conservatism of the TOST. Consequently, the test provides better power compared to the TOST, especially when the variance becomes large. Numerical results also show that our test results in power that is very similar to the other improved tests available in the literature. An advantage of our improved test is that it is just as easy to carry out as the TOST. An example is used to illustrate the new test.     

Testing for Hardy-Weinberg equilibrium in samples with related individuals

When the classical chi(2) goodness-of-fit test for Hardy-Weinberg (HW) equilibrium is used on samples with related individuals, the type I error can be greatly inflated. In particular the test is inappropriate in population isolates where the individuals are related through multiple lines of descent. In this article, we propose a new test for HW (the QL-HW test) suitable for any sample with related individuals, including large inbred pedigrees, provided that their genealogy is known. Performed conditional on the pedigree structure, the QL-HW test detects departures from HW that are not due to the genealogy. Because the computation of the QL-HW test becomes intractable for very polymorphic loci in large inbred pedigrees, a simpler alternative, the GCC-HW test, is also proposed. The statistical properties of the QL-HW and GCC-HW tests are studied through simulations considering a sample of independent nuclear families, a sample of extended outbred genealogies, and samples from the Hutterite population, a North American highly inbred isolate. Finally, the method is used to test a set of 143 biallelic markers spanning 82 genes in this latter population.