Robust Univariate Two-Way Classification

We develop a robust classification procedure, based on TIKU'S (1967, 1980, 1982) MML (modified maximum likelihood) estimators, for classifying an observation in one of the two populations π₁ and π₂ and show that this procedure is superior to the classical and nonparametric procedures.     

A Robust Test for Multi-Sample Location Problems with Unequal Group Variances and Nonnormality

A robust test (to be referred to as M^* test) is proposed for testing equality of several group means without assuming normality and equality of variances. This test statistic is obtained by combining Tiku's MML robust procedure with the James statistic. Monte Carlo simulation studies indicate that the M^* test is more powerful than the Welch test, the James test, and the tests based on Huber's M-estimators over a wide range of nonnormal universes. It is also more powerful than the Brown and Forsythe test under most of nonnormal distributions and has substantially the same power as the Brown and Forsythe test under normal distribution. Comparing with Tan-Tabatabai test, M^* is almost as powerful as Tan-Tabatabai test.     

Robust alternatives to the F-Test in mixed linear models based on MM-estimates

Mixed linear models are commonly used to analyze data in many settings. These models are generally fitted by means of (restricted) maximum likelihood techniques relying heavily on normality. The sensitivity of the resulting estimators and related tests to this underlying assumption has been identified as a weakness that can even lead to wrong interpretations. Very recently a highly robust estimator based on a scale estimate, that is, an S-estimator, has been proposed for general mixed linear models. It has the advantage of being easy to compute and allows the computation of a robust score test. However, this proposal cannot be used to define a likelihood ratio type test that is certainly the most direct route to robustify an F-test. As the latter is usually a key tool of hypothesis testing in mixed linear models, we propose two new robust estimators that allow the desired extension. They also lead to resistant Wald-type tests useful for testing contrasts and covariate effects. We study their properties theoretically and by means of simulations. The analysis of a real data set illustrates the advantage of the new approach in the presence of outlying observations.     

Analysis of covariance with pre-treatment measurements in randomized trials: comparison of equal and unequal slopes

In randomized trials, an analysis of covariance (ANCOVA) is often used to analyze post-treatment measurements with pre-treatment measurements as a covariate to compare two treatment groups. Random allocation guarantees only equal variances of pre-treatment measurements. We hence consider data with unequal covariances and variances of post-treatment measurements without assuming normality. Recently, we showed that the actual type I error rate of the usual ANCOVA assuming equal slopes and equal residual variances is asymptotically at a nominal level under equal sample sizes, and that of the ANCOVA with unequal variances is asymptotically at a nominal level, even under unequal sample sizes. In this paper, we investigated the asymptotic properties of the ANCOVA with unequal slopes for such data. The estimators of the treatment effect at the observed mean are identical between equal and unequal variance assumptions, and these are asymptotically normal estimators for the treatment effect at the true mean. However, the variances of these estimators based on standard formulas are biased, and the actual type I error rates are not at a nominal level, irrespective of variance assumptions. In equal sample sizes, the efficiency of the usual ANCOVA assuming equal slopes and equal variances is asymptotically the same as those of the ANCOVA with unequal slopes and higher than that of the ANCOVA with equal slopes and unequal variances. Therefore, the use of the usual ANCOVA is appropriate in equal sample sizes.     

Longitudinal analysis of pre- and post-treatment measurements with equal baseline assumptions in randomized trials

For continuous variables of randomized controlled trials, recently, longitudinal analysis of pre- and posttreatment measurements as bivariate responses is one of analytical methods to compare two treatment groups. Under random allocation, means and variances of pretreatment measurements are expected to be equal between groups, but covariances and posttreatment variances are not. Under random allocation with unequal covariances and posttreatment variances, we compared asymptotic variances of the treatment effect estimators in three longitudinal models. The data-generating model has equal baseline means and variances, and unequal covariances and posttreatment variances. The model with equal baseline means and unequal variance–covariance matrices has a redundant parameter. In large sample sizes, these two models keep a nominal type I error rate and have high efficiency. The model with equal baseline means and equal variance–covariance matrices wrongly assumes equal covariances and posttreatment variances. Only under equal sample sizes, this model keeps a nominal type I error rate. This model has the same high efficiency with the data-generating model under equal sample sizes. In conclusion, longitudinal analysis with equal baseline means performed well in large sample sizes. We also compared asymptotic properties of longitudinal models with those of the analysis of covariance (ANCOVA) and t-test.     

A mixture-model approach for parallel testing for unequal variances

Testing for unequal variances is usually performed in order to check the validity of the assumptions that underlie standard tests for differences between means (the t-test and anova). However, existing methods for testing for unequal variances (Levene's test and Bartlett's test) are notoriously non-robust to normality assumptions, especially for small sample sizes. Moreover, although these methods were designed to deal with one hypothesis at a time, modern applications (such as to microarrays and fMRI experiments) often involve parallel testing over a large number of levels (genes or voxels). Moreover, in these settings a shift in variance may be biologically relevant, perhaps even more so than a change in the mean. This paper proposes a parsimonious model for parallel testing of the equal variance hypothesis. It is designed to work well when the number of tests is large; typically much larger than the sample sizes. The tests are implemented using an empirical Bayes estimation procedure which `borrows information' across levels. The method is shown to be quite robust to deviations from normality, and to substantially increase the power to detect differences in variance over the more traditional approaches even when the normality assumption is valid.     

On the estimation of average treatment effects with right-censored time to event outcome and competing risks

We are interested in the estimation of average treatment effects based on right-censored data of an observational study. We focus on causal inference of differences between t-year absolute event risks in a situation with competing risks. We derive doubly robust estimation equations and implement estimators for the nuisance parameters based on working regression models for the outcome, censoring, and treatment distribution conditional on auxiliary baseline covariates. We use the functional delta method to show that these estimators are regular asymptotically linear estimators and estimate their variances based on estimates of their influence functions. In empirical studies, we assess the robustness of the estimators and the coverage of confidence intervals. The methods are further illustrated using data from a Danish registry study.     

To permute or not to permute

Permutation test is a popular technique for testing a hypothesis of no effect, when the distribution of the test statistic is unknown. To test the equality of two means, a permutation test might use a test statistic which is the difference of the two sample means in the univariate case. In the multivariate case, it might use a test statistic which is the maximum of the univariate test statistics. A permutation test then estimates the null distribution of the test statistic by permuting the observations between the two samples. We will show that, for such tests, if the two distributions are not identical (as for example when they have unequal variances, correlations or skewness), then a permutation test for equality of means based on difference of sample means can have an inflated Type I error rate even when the means are equal. Our results illustrate permutation testing should be confined to testing for non-identical distributions. CONTACT: calian@raunvis.hi.is.     

Foraging preferences: Response to reward skew

Recent theoretical and empirical work on foraging behaviour suggests that animals may respond to both the means and variances in benefits associated with available resources. We attempt to extend this analysis by asking if reward skew (third moment about the mean) might influence preference when two options have equal means and equal variances. We examine how minimizing the probability of starvation might induce response to skew. In the Appendix we develop an expected ‘fitness’ model which follows from economic theory and indicates more general conditions concerning responses to skew. We also report experiments involving foraging white-crowned sparrows (Zonotrichia leucophrys). Under conditions where positive skew should be favoured, the birds' behaviour supports the prediction. However, their response to skew is not as strong as responses to variance noted in the same individuals.     

Blinded and unblinded sample size reestimation in crossover trials balanced for period

The determination of the sample size required by a crossover trial typically depends on the specification of one or more variance components. Uncertainty about the value of these parameters at the design stage means that there is often a risk a trial may be under‐ or overpowered. For many study designs, this problem has been addressed by considering adaptive design methodology that allows for the re‐estimation of the required sample size during a trial. Here, we propose and compare several approaches for this in multitreatment crossover trials. Specifically, regulators favor reestimation procedures to maintain the blinding of the treatment allocations. We therefore develop blinded estimators for the within and between person variances, following simple or block randomization. We demonstrate that, provided an equal number of patients are allocated to sequences that are balanced for period, the proposed estimators following block randomization are unbiased. We further provide a formula for the bias of the estimators following simple randomization. The performance of these procedures, along with that of an unblinded approach, is then examined utilizing three motivating examples, including one based on a recently completed four‐treatment four‐period crossover trial. Simulation results show that the performance of the proposed blinded procedures is in many cases similar to that of the unblinded approach, and thus they are an attractive alternative.     

A Monte Carlo method for Bayesian analysis of linkage between single markers and quantitative trait loci. I. Methodology

A Bayesian approach to the statistical mapping of Quantitative Trait Loci (QTLs) using single markers was implemented via Markov Chain Monte Carlo (MCMC) algorithms for parameter estimation and hypothesis testing. Parameter estimators were marginal posterior means computed using a Gibbs sampler with data augmentation. Variables sampled included the augmented data (marker-QTL genotypes, polygenic effects), an indicator variable for linkage, and the parameters (allele frequency, QTL substitution effect, recombination rate, polygenic and residual variances). Several MCMC algorithms were derived for computing Bayesian tests of linkage, which consisted of the marginal posterior probability of linkage and the marginal likelihood of the QTL variance associated with the marker.     

Tests Against Qualitative Interaction: Exact Critical Values and Robust Tests

Consider a study to evaluate treatment A with a placebo in two or more groups of patients. If treatment A is beneficial to one group of patients and harmful to another, then we say that there is qualitative interaction or crossover interaction between patient groups and the treatments. Gail and Simon (1985, Biometrics 41, 361-372) developed a large-sample procedure for this testing problem. Their test has received favorable coverage in the literature. In this article, we obtain corresponding exact finite sample results for normal error distribution and provide a table of critical values. The test statistic is similar to the familiar F-ratio, and its p-value is equal to a weighted sum of tail areas of F-distributions. The computations to implement this are simple. A simulation study shows that the exact critical values provided here for normal error distribution are preferable to the asymptotic critical values for a wide range of error distributions. We also develop tests that are power robust against long-tailed error distributions. Our robust test uses M-estimators instead of the least squares estimators. We show that the efficiency robustness of the M-estimator translates to power robustness of the corresponding test. Therefore, our robust tests are better if outliers are expected. A simulation study illustrates the substantial power advantages of our robust tests.     

Improving data interpretation of fragmentary data-sets on invertebrate dispersal with permutation tests

Biological data often tend to have heterogeneous, discontinuous non-normal distributions. Statistical non-parametric tests, like the Mann–Whitney U-test or the extension for more than two samples, the Kruskal–Wallis test, are often used in these cases, although they assume certain preconditions which are often ignored. We developed a permutation test procedure that uses the ratio of the interquartile distances and the median differences of the original non-classified data to assess the properties of the real distribution more appropriately than the classical methods. We used this test on a heterogeneous, skewed biological data set on invertebrate dispersal and showed how different the reactions of the Kruskal–Wallis test and the permutation approach are. We then evaluated the new testing procedure with reproducible data that were generated from the normal distribution. Here, we tested the influence of four different experimental trials on the new testing procedure in comparison to the Kruskal–Wallis test. These trials showed the impact of data that were varying in terms of (a) negative correlation between variances and means of the samples, (b) changing variances that were not correlated with the means of the samples, (c) constant variances and means, but different sample sizes and in trials (d) we evaluated the testing power of the new procedure. Due to the different test statistics, the permutation test reacted more sensibly to the data presented in trials (a) and c) and non-uniformly in trial (b). In the evaluation of the testing power, no significant differences between the Kruskal–Wallis test and the new permutation testing procedure could be detected. We consider this test to be an alternative for working on heterogeneous data where the preconditions of the classical non-parametric tests are not met.     

Tests for Homogeneity of Variances Using Robust Weighted Likelihood Estimates

The classical normal-theory tests for testing the null hypothesis of common variance and the classical estimates of scale have long been known to be quite nonrobust to even mild deviations from normality assumptions for moderate sample sizes. Levene (1960) suggested a one-way ANOVA type statistic as a robust test. Brown and Forsythe (1974) considered a modified version of Levene's test by replacing the sample means with sample medians as estimates of population locations, and their test is computationally the simplest among the three tests recommended by Conover , Johnson , and Johnson (1981) in terms of robustness and power. In this paper a new robust and powerful test for homogeneity of variances is proposed based on a modification of Levene's test using the weighted likelihood estimates (Markatou , Basu , and Lindsay , 1996) of the population means. For two and three populations the proposed test using the Hellinger distance based weighted likelihood estimates is observed to achieve better empirical level and power than Brown-Forsythe's test in symmetric distributions having a thicker tail than the normal, and higher empirical power in skew distributions under the use of F distribution critical values.     

The Repertoire of Plant Cells

Cryptic (camouflaged) prey often seek out backgrounds that match their coloration, and when at rest adopt an attitude that makes their crypsis most effective. We suggest a simple method for investigating the adaptive significance of such orientation. We used flat discs of pastry as the ‘prey’, either plain white or painted with a central black stripe, and wild garden birds as the predators. In the eight main experiments the backgrounds were white wooden boards painted with black parallel stripes of the same width as the stripes on the prey. In each experiment we presented equal numbers of two (of seven) ‘treatments’ of prey. The selection resulting from the combined predation by the birds confirmed the advantages of resting on a matching background and in the ‘correct’ orientation. We suggest that the technique can be developed further to explore the adaptive significance of background matching     

A covariance estimator for GEE with improved small-sample properties

In this paper, we propose an alternative covariance estimator to the robust covariance estimator of generalized estimating equations (GEE). Hypothesis tests using the robust covariance estimator can have inflated size when the number of independent clusters is small. Resampling methods, such as the jackknife and bootstrap, have been suggested for covariance estimation when the number of clusters is small. A drawback of the resampling methods when the response is binary is that the methods can break down when the number of subjects is small due to zero or near-zero cell counts caused by resampling. We propose a bias-corrected covariance estimator that avoids this problem. In a small simulation study, we compare the bias-corrected covariance estimator to the robust and jackknife covariance estimators for binary responses for situations involving 10-40 subjects with equal and unequal cluster sizes of 16-64 observations. The bias-corrected covariance estimator gave tests with sizes close to the nominal level even when the number of subjects was 10 and cluster sizes were unequal, whereas the robust and jackknife covariance estimators gave tests with sizes that could be 2-3 times the nominal level. The methods are illustrated using data from a randomized clinical trial on treatment for bone loss in subjects with periodontal disease.     

Statistical Analyses of Twins Birth Data from a Nigerian Sample

In this paper, both the linear logistic model and its analogous linear model (weighted least squares) are fitted to twin birth data from a Nigerian sample. The logits of the observed perinatal mortality rates are fitted against the birthweight of breech infants for both first born and second born twins. Results show that a quadratic response model fits the data very well, and that the WLS procedure gives a better fit. The results further show that breech births necessarily result in 7.7% and 11.2% mortality rates respectively for first born and second born twins. The results of a log-linear model analysis on the 2×2×4 contingency table formed by the factors, ‘mode of birth’, and twins with the response variable ‘Apgar Scores’-an index of morbidity-also show that both factors play significant roles in explaining the variation of the response variable. It was found that the most important factor effecting the ability of a twin infant to survive as measured by the Apgar scores is the variable ‘mode of birth’ (vertex or breech). In other words, a vertex first twin infant has the best survival chance.     

Joint evolution of differential seed dispersal and self‐fertilization

Differential seed dispersal, in which selfed and outcrossed seeds possess different dispersal propensities, represents a potentially important individual‐level association. A variety of traits can mediate differential seed dispersal, including inflorescence and seed size variation. However, how natural selection shapes such associations is poorly known. Here, we developed theoretical models for the evolution of mating system and differential seed dispersal in metapopulations, incorporating heterogeneous pollination, dispersal cost, cost of outcrossing and environment‐dependent inbreeding depression. We considered three models. In the ‘fixed dispersal model’, only selfing rate is allowed to evolve. In the ‘fixed selfing model’, in which selfing is fixed but differential seed dispersal can evolve, we showed that natural selection favours a higher, equal or lower dispersal rate for selfed seeds to that for outcrossed seeds. However, in the ‘joint evolution model’, in which selfing and dispersal can evolve together, evolution necessarily leads to higher or equal dispersal rate for selfed seeds compared to that for outcrossed. Further comparison revealed that outcrossed seed dispersal is selected against by the evolution of mixed mating or selfing, whereas the evolution of selfed seed dispersal undergoes independent processes. We discuss the adaptive significance and constraints for mating system/dispersal association.     

Monogamy Has a Fixation Advantage Based on Fitness Variance in an Ideal Promiscuity Group

We consider an ideal promiscuity group of females, which implies that all males have the same average mating success. If females have concealed ovulation, then the males’ paternity chances are equal. We find that male-based monogamy will be fixed in females’ promiscuity group when the stochastic Darwinian selection is described by a Markov chain. We point out that in huge populations the relative advantage (difference between average fitness of different strategies) determines primarily the end of evolution; in the case of neutrality (means are equal) the smallest variance guarantees fixation (absorption) advantage; when the means and variances are the same, then the higher third moment determines which types will be fixed in the Markov chains.     

FORAGING TRAIT (CO)VARIANCES IN STICKLEBACK EVOLVE DETERMINISTICALLY AND DO NOT PREDICT TRAJECTORIES OF ADAPTIVE DIVERSIFICATION

How does natural selection shape the structure of variance and covariance among multiple traits, and how do (co)variances influence trajectories of adaptive diversification? We investigate these pivotal but open questions by comparing phenotypic (co)variances among multiple morphological traits across 18 derived lake‐dwelling populations of threespine stickleback, and their marine ancestor. Divergence in (co)variance structure among populations is striking and primarily attributable to shifts in the variance of a single key foraging trait (gill raker length). We then relate this divergence to an ecological selection proxy, to population divergence in trait means, and to the magnitude of sexual dimorphism within populations. This allows us to infer that evolution in (co)variances is linked to variation among habitats in the strength of resource‐mediated disruptive selection. We further find that adaptive diversification in trait means among populations has primarily involved shifts in gill raker length. The direction of evolutionary trajectories is unrelated to the major axes of ancestral trait (co)variance. Our study demonstrates that natural selection drives both means and (co)variances deterministically in stickleback, and strongly challenges the view that the (co)variance structure biases the direction of adaptive diversification predictably even over moderate time spans.