Search

Generalized linear models often have high-dimensional nuisance parameters, as seen in applications such as testing gene-environment interactions or gene-gene interactions. In these scenarios, it is essential to test the significance of a high-dimensional subvector of the model's coefficients. Although some existing methods can tackle this problem, they often rely on the bootstrap to approximate the asymptotic distribution of the test statistic, and are thus computationally expensive. Here, we propose a computationally efficient test with a closed-form limiting distribution, which allows the parameter being tested to be either sparse or dense. We show that, under certain regularity conditions, the Type-I error of the proposed method is asymptotically correct, and we establish its power under high-dimensional alternatives. Extensive simulations demonstrate the good performance of the proposed test and its robustness when certain sparsity assumptions are violated. We also apply the proposed method to Chinese famine sample data in order to show its performance when testing the significance of gene-environment interactions. [ABSTRACT FROM AUTHOR]

We consider the problem of testing for the presence of linear relationships between large sets of random variables based on a postselection inference approach to canonical correlation analysis. The challenge is to adjust for the selection of subsets of variables having linear combinations with maximal sample correlation. To this end, we construct a stabilized one-step estimator of the Euclidean norm of the canonical correlations maximized over subsets of variables of prespecified cardinality. This estimator is shown to be consistent for its target parameter and asymptotically normal, provided the dimensions of the variables do not grow too quickly with sample size. We also develop a greedy search algorithm to accurately compute the estimator, leading to a computationally tractable omnibus test for the global null hypothesis that there are no linear relationships between any subsets of variables having the prespecified cardinality. We further develop a confidence interval that takes the variable selection into account. [ABSTRACT FROM AUTHOR]

Identifying dependency in multivariate data is a common inference task that arises in numerous applications. However, existing nonparametric independence tests typically require computation that scales at least quadratically with the sample size, making it difficult to apply them in the presence of massive sample sizes. Moreover, resampling is usually necessary to evaluate the statistical significance of the resulting test statistics at finite sample sizes, further worsening the computational burden. We introduce a scalable, resampling-free approach to testing the independence between two random vectors by breaking down the task into simple univariate tests of independence on a collection of |$2\times 2$| contingency tables constructed through sequential coarse-to-fine discretization of the sample , transforming the inference task into a multiple testing problem that can be completed with almost linear complexity with respect to the sample size. To address increasing dimensionality, we introduce a coarse-to-fine sequential adaptive procedure that exploits the spatial features of dependency structures. We derive a finite-sample theory that guarantees the inferential validity of our adaptive procedure at any given sample size. We show that our approach can achieve strong control of the level of the testing procedure at any sample size without resampling or asymptotic approximation and establish its large-sample consistency. We demonstrate through an extensive simulation study its substantial computational advantage in comparison to existing approaches while achieving robust statistical power under various dependency scenarios, and illustrate how its divide-and-conquer nature can be exploited to not just test independence, but to learn the nature of the underlying dependency. Finally, we demonstrate the use of our method through analysing a dataset from a flow cytometry experiment. [ABSTRACT FROM AUTHOR]

Probability forecasts for binary events play a central role in many applications. Their quality is commonly assessed with proper scoring rules, which assign forecasts numerical scores such that a correct forecast achieves a minimal expected score. In this paper, we construct e-values for testing the statistical significance of score differences of competing forecasts in sequential settings. E-values have been proposed as an alternative to |$p$| -values for hypothesis testing, and they can easily be transformed into conservative |$p$| -values by taking the multiplicative inverse. The e-values proposed in this article are valid in finite samples without any assumptions on the data-generating processes. They also allow optional stopping, so a forecast user may decide to interrupt evaluation, taking into account the available data at any time, and still draw statistically valid inference, which is generally not true for classical |$p$| -value-based tests. In a case study on post-processing of precipitation forecasts, state-of-the-art forecast dominance tests and e-values lead to the same conclusions. [ABSTRACT FROM AUTHOR]

An important and often discussed research problem in statistics is how to compare several populations; examples arise in medical science, engineering, finance and other fields. Often population means or medians are compared. However, one population may have a higher mean income, for example, because of a small number of super-rich individuals; the mean therefore may not reflect the wealth of the general population. Instead, an index of the degree of stochastic dominance of one population over another would better reflect their relative wealth. Currently, we can estimate such an index under restrictive conditions, but there is no generic estimator with a known asymptotic distribution. In this paper, we suggest linking the populations via the density ratio model. Under this model, we develop an empirical likelihood estimator and establish its asymptotic normality. In addition, we improve the estimation efficiency by examining the similarities between the populations. Furthermore, we provide a valid bootstrap method for hypothesis testing and the construction of confidence intervals. Simulation experiments show that the proposed estimator substantially improves the estimation efficiency and power of the test, and leads to confidence intervals with satisfactorily precise coverage probabilities. It is also robust with respect to mild model misspecification. Two examples are given to demonstrate the usefulness of both the method and the concept. [ABSTRACT FROM AUTHOR]

Replicability analysis aims to identify the overlapping signals across independent studies that examine the same features. For this purpose we develop hypothesis testing procedures that first select the promising features from each of two studies separately. Only those features selected in both studies are then tested. The proposed procedures have theoretical guarantees regarding their control of the familywise error rate or false discovery rate on the replicability claims. They can also be used for signal discovery in each study separately, with the desired error control. Their power for detecting truly replicable findings is compared to alternatives. We illustrate the procedures on behavioural genetics data. [ABSTRACT FROM AUTHOR]

Combining p-values from independent statistical tests is a popular approach to meta-analysis, particularly when the data underlying the tests are either no longer available or are difficult to combine. Numerous p-value combination methods appear in the literature, each with different statistical properties, yet often the final choice used in a meta-analysis can seem arbitrary, as if all effort has been expended in building the models that gave rise to the p-values. Birnbaum (1954) showed that any reasonable p-value combiner must be optimal against some alternative hypothesis. Starting from this perspective and recasting each method of combining p-values as a likelihood ratio test, we present theoretical results for some standard combiners that provide guidance on how a powerful combiner might be chosen in practice. [ABSTRACT FROM AUTHOR]

Fisher randomization tests for Neyman's null hypothesis of no average treatment effect are considered in a finite-population setting associated with completely randomized experiments involving more than two treatments. The consequences of using the F statistic to conduct such a test are examined, and we argue that under treatment effect heterogeneity, use of the F statistic in the Fisher randomization test can severely inflate theType I error under Neyman's null hypothesis. We propose to use an alternative test statistic, derive its asymptotic distributions under Fisher's and Neyman's null hypotheses, and demonstrate its advantages through simulations. [ABSTRACT FROM AUTHOR]

Separability is a common simplifying assumption on the covariance structure of spatiotemporal functional data. We present three tests of separability, one a functional extension of the Monte Carlo likelihood method of Mitchell et al. (2006) and two based on quadratic forms. Our tests are based on asymptotic distributions of maximum likelihood estimators and do not require Monte Carlo simulation. The main theoretical contribution of this paper is the specification of the joint asymptotic distribution of these estimators, which can be used to derive many other tests. The main methodological finding is that one of the quadratic form methods, which we call a norm approach, emerges as a clear winner in terms of finite-sample performance in nearly every setting we considered. This approach focuses directly on the Frobenius distance between the spatiotemporal covariance function and its separable approximation. We demonstrate the efficacy of our methods via simulations and application to Irish wind data. [ABSTRACT FROM AUTHOR]

Several two-sample tests for high-dimensional data have been proposed recently, but they are powerful only against certain alternative hypotheses. In practice, since the true alternative hypothesis is unknown, it is unclear how to choose a powerful test. We propose an adaptive test that maintains high power across a wide range of situations and study its asymptotic properties. Its finite-sample performance is compared with that of existing tests. We apply it and other tests to detect possible associations between bipolar disease and a large number of single nucleotide polymorphisms on each chromosome based on data from a genome-wide association study. Numerical studies demonstrate the superior performance and high power of the proposed test across a wide spectrum of applications. [ABSTRACT FROM AUTHOR]

We address the problem of testing for a parametric function of fixed effects in mixed models. We propose a test based on the distance between two empirical error distribution functions, which are constructed from residuals calculated under the opposing hypotheses. The proposed test statistic has power against all alternatives, and its asymptotic distribution is derived. A simulation study shows that the test outperforms others in the literature. The test is applied to longitudinal data from an AIDS clinical trial and a growth study. [ABSTRACT FROM AUTHOR]

We consider large scale multiple testing for data that have locally clustered signals. With this structure, we apply techniques from changepoint analysis and propose a boundary detection algorithm so that the clustering information can be utilized. Consequently the precision of the multiple testing procedure is substantially improved. We study tests with independent as well as dependent p-values. Monte Carlo simulations suggest that the methods perform well with realistic sample sizes and show improved detection ability compared with competing methods. Our procedure is applied to a genome-wide association dataset of blood lipids. [ABSTRACT FROM AUTHOR]

A framework for classification is developed with a notion of confidence. In this framework, a classifier consists of two tolerance regions in the predictor space, with a specified coverage level for each class. The classifier also produces an ambiguous region where the classification needs further investigation. Theoretical analysis reveals interesting structures of the confidence-ambiguity trade-off, and the optimal solution is characterized by extending the Neyman–Pearson lemma. We provide general estimating procedures, along with rates of convergence, based on estimates of the conditional probabilities. The method can be easily implemented with good robustness, as illustrated through theory, simulation and a data example. [ABSTRACT FROM PUBLISHER]

In this paper we derive a new p-value based multiple testing procedure that improves upon the Hommel procedure by gaining power as well as having a simpler step-up structure similar to the Hochberg procedure. The key to this improvement is that the Hommel procedure can be improved by a consonant procedure. Exact critical constants of this new procedure can be numerically determined. The zeroth-order approximations to the exact critical constants, albeit slightly conservative, are simple to use and need no tabling, and hence are recommended in practice. The proposed procedure is shown to control the familywise error rate under independence among the p-values. Simulations empirically demonstrate familywise error rate control under positive and negative dependence. Power superiority of the proposed procedure over competing ones is also empirically demonstrated. Illustrative examples are given. [ABSTRACT FROM PUBLISHER]

The higher criticism test is effective for testing a joint null hypothesis against a sparse alternative, e.g., for testing the effect of a gene or genetic pathway that consists of d genetic markers. Accurate p-value calculations for the higher criticism test based on the asymptotic distribution require a very large d, which is not the case for the number of genetic variants in a gene or a pathway. In this paper we propose an analytical method for accurately computing the p-value of the higher criticism test for finite-d problems. Unlike previous treatments, this method does not rely on asymptotics in d or on simulation, and is exact for arbitrary d when the test statistics are normally distributed. The method is particularly computationally advantageous when d is not large. We illustrate the proposed method with a case-control genome-wide association study of lung cancer and compare its power with competing methods through simulations. [ABSTRACT FROM PUBLISHER]

When the multiple correlation coefficient is used to measure how strongly a given variable can be linearly associated with a set of covariates, it suffers from an upward bias that cannot be ignored in the presence of a moderately high dimensional covariate. Under an independent component model, we derive an asymptotic approximation to the distribution of the squared multiple correlation coefficient that depends on a simple correction factor. We show that this approximation enables us to construct reliable confidence intervals on the population coefficient even when the ratio of the dimension to the sample size is close to unity and the variables are non-Gaussian. [ABSTRACT FROM PUBLISHER]

Many statistical analysis procedures require a good estimator for a high-dimensional covariance matrix or its inverse, the precision matrix. When the precision matrix is banded, the Cholesky-based method often yields a good estimator of the precision matrix. One important aspect of this method is determination of the band size of the precision matrix. In practice, crossvalidation is commonly used; however, we show that crossvalidation not only is computationally intensive but can be very unstable. In this paper, we propose a new hypothesis testing procedure to determine the band size in high dimensions. Our proposed test statistic is shown to be asymptotically normal under the null hypothesis, and its theoretical power is studied. Numerical examples demonstrate the effectiveness of our testing procedure. [ABSTRACT FROM PUBLISHER]

The problem of testing smooth components of an extended generalized additive model for equality to zero is considered. Confidence intervals for such components exhibit good across-the-function coverage probabilities if based on the approximate result , where f is the vector of evaluated values for the smooth component of interest and V f is the covariance matrix for f according to the Bayesian view of the smoothing process. Based on this result, a Wald-type test of f=0 is proposed. It is shown that care must be taken in selecting the rank used in the test statistic. The method complements previous work by extending applicability beyond the Gaussian case, while considering tests of zero effect rather than testing the parametric hypothesis given by the null space of the component’s smoothing penalty. The proposed p-values are routine and efficient to compute from a fitted model, without requiring extra model fits or null distribution simulation. [ABSTRACT FROM PUBLISHER]

Merging data from multiple studies has been widely adopted in biomedical research. In this paper, we consider two major issues related to merging longitudinal datasets. We first develop a rigorous hypothesis testing procedure to assess the validity of data merging, and then propose a flexible joint estimation procedure that enables us to analyse merged data and to account for different within-subject correlations and follow-up schedules in different studies. We establish large sample properties for the proposed procedures. We compare our method with meta analysis and generalized estimating equations and show that our test provides robust control of Type I error against both misspecification of working correlation structures and heterogeneous dispersion parameters. Our joint estimating procedure leads to an improvement in estimation efficiency on all regression coefficients after data merging is validated. [ABSTRACT FROM PUBLISHER]

The false discovery rate is a criterion for controlling Type I error in simultaneous testing of multiple hypotheses. For scanning statistics, due to local dependence, clusters of neighbouring hypotheses are likely to be rejected together. In such situations, it is more intuitive and informative to group neighbouring rejections together and count them as a single discovery, with the false discovery rate defined as the proportion of clusters that are falsely declared among all declared clusters. Assuming that the number of false discoveries, under this broader definition of a discovery, is approximately Poisson and independent of the number of true discoveries, we examine approaches for estimating and controlling the false discovery rate, and provide examples from biological applications. [ABSTRACT FROM PUBLISHER]

An integrated likelihood depends only on the parameter of interest and the data, so it can be used as a standard likelihood function for likelihood-based inference. In this paper, the higher-order asymptotic properties of the signed integrated likelihood ratio statistic for a scalar parameter of interest are considered. These results are used to construct a modified integrated likelihood ratio statistic and to suggest a class of prior densities to use in forming the integrated likelihood. The properties of the integrated likelihood ratio statistic are compared to those of the standard likelihood ratio statistic. Several examples show that the integrated likelihood ratio statistic can be a useful alternative to the standard likelihood ratio statistic. [ABSTRACT FROM PUBLISHER]

In the first stage of a two-stage, drop-the-losers design, a candidate for the best treatment is selected. At the second stage, additional observations are collected to decide whether the candidate is actually better than the control. The design also allows the investigator to stop the trial for ethical reasons at the end of the first stage if there is already strong evidence of futility or superiority. Two types of tests have recently been developed, one based on the combined means and the other based on the combined p-values, but corresponding interval estimators are unavailable except in special cases. The problem is that, in most cases, the interval estimators depend on the mean configuration of all treatments in the first stage, which is unknown. In this paper, we prove a basic stochastic ordering lemma that enables us to bridge the gap between hypothesis testing and interval estimation. The proposed confidence intervals achieve the nominal confidence level in certain special cases. Simulations show that decisions based on our intervals are usually more powerful than those based on existing methods. [ABSTRACT FROM PUBLISHER]

This paper discusses copula model selection procedures and goodness-of-fit tests under censoring. The proposed methodology is based on a comparison of nonparametric and model-based estimators of the probability integral transformation, K. New weighted estimators for K are introduced. The resulting tests are compared to an existing approach by simulation and illustrated with an example involving bleeding changes in a woman’s reproductive history. [ABSTRACT FROM PUBLISHER]

The covariate-adaptive randomization method was proposed for clinical trials long ago but little theoretical work has been done for statistical inference associated with it. Practitioners often apply test procedures available for simple randomization, which is controversial since procedures valid under simple randomization may not be valid under other randomization schemes. In this paper, we provide some theoretical results for testing hypotheses after covariate-adaptive randomization. We show that one way to obtain a valid test procedure is to use a correct model between outcomes and covariates, including those used in randomization. We also show that the simple two sample t-test, without using any covariate, is conservative under covariate-adaptive biased coin randomization in terms of its Type I error, and that a valid bootstrap t-test can be constructed. The powers of several tests are examined theoretically and empirically. Our study provides guidance for applications and sheds light on further research in this area. [ABSTRACT FROM PUBLISHER]

We propose pseudo-score confidence intervals for parameters in models for discrete data. The confidence interval is obtained by inverting a test that uses a Pearson chi-squared statistic to compare fitted values for the working model with fitted values of the model when a parameter of interest takes various fixed values. For multinomial models, the pseudo-score method simplifies to the score method when the model is saturated and otherwise it is asymptotically equivalent to score and likelihood ratio test-based inferences. For cases in which ordinary score methods are impractical, such as when the likelihood function is not an explicit function of model parameters, the pseudo-score method is feasible. We illustrate the method for four such examples. Generalizations of the method are also presented for future research, including inference for complex sampling designs using a quasilikelihood Pearson statistic that compares fitted values for two models relative to the variance of the observations under the simpler model. [ABSTRACT FROM PUBLISHER]

We consider variable selection in the single-index model. We prove that the popular leave-m-out crossvalidation method has different behaviour in the single-index model from that in linear regression models or nonparametric regression models. A new consistent variable selection method, called separated crossvalidation, is proposed. Further analysis suggests that the method has better finite-sample performance and is computationally easier than leave-m-out crossvalidation. Separated crossvalidation, applied to the Swiss banknotes data and the ozone concentration data, leads to single-index models with selected variables that have better prediction capability than models based on all the covariates. [ABSTRACT FROM AUTHOR]

Although Simes' modification of the Bonferroni procedure tends to perform very well, albeit often being slightly liberal for negatively dependent hypotheses, there are special cases where it fails more dramatically. We prove that these special cases are indeed special, applying only to specific significance levels, and obtain a strong bound on the average deviation of the Simes corrected P-value from the true probability over any interval of P-values. From this, it is argued that Simes' procedure should be expected to perform well except for pathological examples. [ABSTRACT FROM AUTHOR]

We present a method for multiple hypothesis testing that maintains control of the false discovery rate while incorporating prior information about the hypotheses. The prior information takes the form of p-value weights. If the assignment of weights is positively associated with the null hypotheses being false, the procedure improves power, except in cases where power is already near one. Even if the assignment of weights is poor, power is only reduced slightly, as long as the weights are not too large. We also provide a similar method for controlling false discovery exceedance. [ABSTRACT FROM AUTHOR]

The linear step-up multiple testing procedure controls the false discovery rate at the desired level q for independent and positively dependent test statistics. When all null hypotheses are true, and the test statistics are independent and continuous, the bound is sharp. When some of the null hypotheses are not true, the procedure is conservative by a factor which is the proportion m0/m of the true null hypotheses among the hypotheses. We provide a new two-stage procedure in which the linear step-up procedure is used in stage one to estimate m0, providing a new level q′ which is used in the linear step-up procedure in the second stage. We prove that a general form of the two-stage procedure controls the false discovery rate at the desired level q. This framework enables us to study analytically the properties of other procedures that exist in the literature. A simulation study is presented that shows that two-stage adaptive procedures improve in power over the original procedure, mainly because they provide tighter control of the false discovery rate. We further study the performance of the current suggestions, some variations of the procedures, and previous suggestions, in the case where the test statistics are positively dependent, a case for which the original procedure controls the false discovery rate. In the setting studied here the newly proposed two-stage procedure is the only one that controls the false discovery rate. The procedures are illustrated with two examples of biological importance. [ABSTRACT FROM AUTHOR]

In this paper we evaluate the characteristics observed both on a crime sample and on individuals included in a database to assess the probability of alternative hypotheses concerning identification. The problem is first addressed by considering a generic characteristic and we demonstrate the problem via a computationally efficient Bayesian network. Then we turn our attention to a heritable DNA trait to show how to evaluate the hypotheses that some individuals, genetically related to the members of the database, are the donors of the crime sample. Then the network is extended to cope with many loci. Applications of the method are provided as well as details of computational requirements. [ABSTRACT FROM AUTHOR]

We propose probabilistic lower bounds for the number of false null hypotheses when testing multiple hypotheses of association simultaneously. The bounds are valid under general and unknown dependence structures between the test statistics. The power of the proposed estimator to detect the full proportion of false null hypotheses is discussed and compared to other estimators. The proposed estimator is shown to deliver a tight probabilistic lower bound for the number of false null hypotheses in a multiple testing situation even under strong dependence between test statistics. [ABSTRACT FROM AUTHOR]

Typical alternative hypotheses in the analysis of residuals of a standard regression model are considered, and for each one a Bayesian diagnostic based on a symmetric form of the Kullback–Leibler divergence is determined. The results include an explicit expression for the diagnostic when the alternative hypothesis is that the errors are generated by an unknown distribution function with a Dirichlet process prior. This expression is immediately interpretable, exactly computable and endowed with important asymptotic connections. A linear approximation of the diagnostic reveals close links with the class of Lagrange multiplier test statistics. When the alternative hypothesis is that the errors are generated by an autoregressive process the linear approximation is proportional to the Box–Pierce statistic or to the Ljung–Box statistic, according to the characteristics of the prior, if the observations have zero mean; it depends on the Durbin–Watson statistic if the errors are first-order autoregressive, and it is related to the Cliff–Ord statistic if they are generated by a first-order spatial autoregression. The sensitivity to the prior of the diagnostic and of its linear approximation is also discussed. [ABSTRACT FROM AUTHOR]

We consider a very general class of empirical discrepancy statistics that includes the Cressie–Read discrepancy statistics and, in particular, the empirical likelihood ratio statistic. Higher-order asymptotics for expected lengths of associated confidence intervals are investigated. An explicit formula is worked out and its use for comparative purposes is discussed. It is seen that the empirical likelihood ratio statistic, which enjoys interesting second-order power properties, loses much of its edge under the present criterion. [ABSTRACT FROM AUTHOR]

We use the empirical likelihood method to derive a test and thus a confidence interval based on the rank estimators of the regression coefficient in the accelerated failure time model. Standard chi-squared distributions are used to calculate the p-value and to construct the confidence interval. Simulations and examples show that the chi-squared approximation to the distribution of the log empirical likelihood ratio performs well, and has some advantages over the existing methods. [ABSTRACT FROM AUTHOR]

We study methods for constructing confidence intervals and confidence bands for estimators of receiver operating characteristics. Particular emphasis is placed on the way in which smoothing should be implemented, when estimating either the characteristic itself or its variance. We show that substantial undersmoothing is necessary if coverage properties are not to be impaired. A theoretical analysis of the problem suggests an empirical, plug‐in rule for bandwidth choice, optimising the coverage accuracy of interval estimators. The performance of this approach is explored. Our preferred technique is based on asymptotic approximation, rather than a more sophisticated approach using the bootstrap, since the latter requires a multiplicity of smoothing parameters all of which must be chosen in nonstandard ways. It is shown that the asymptotic method can give very good performance. [ABSTRACT FROM AUTHOR]

A multivariate Gaussian graphical Markov model for an undirected graph G, also called a covariance selection model or concentration graph model, is defined in terms of the Markov properties, i.e. conditional independences associated with G, which in turn are equivalent to specified zeros among the set of pairwise partial correlation coefficients. By means of Fisher's z‐transformation and Šidák's correlation inequality, conservative simultaneous confidence intervals for the entire set of partial correlations can be obtained, leading to a simple method for model selection that controls the overall error rate for incorrect edge inclusion. The simultaneous p‐values corresponding to the partial correlations are partitioned into three disjoint sets, a significant set S, an indeterminate set I and a nonsignificant set N. Our model selection method selects two graphs, a graph ĜSI whose edges correspond to the set S∪I, and a more conservative graph ĜS whose edges correspond to S only. Similar considerations apply to covariance graph models, which are defined in terms of marginal independence rather than conditional independence. The method is applied to some well‐known examples and to simulated data. [ABSTRACT FROM AUTHOR]

This paper studies properties of ordinary and generalised least squares estimators in a simple linear regression with stationary autocorrelated errors. Explicit expressions for the variances of the regression parameter estimators are derived for some common time series autocorrelation structures, including a first‐order autoregression and general moving averages. Applications of the results include confidence intervals and an example where the variance of the trend slope estimator does not increase with increasing autocorrelation. [ABSTRACT FROM AUTHOR]

An unbiased test for the appropriateness of the simple linear regression model is presented. The null hypothesis is that the underlying regression function is indeed a line, and the alternative is that it is convex. The exact distribution for a likelihood ratio test statistic is that of a mixture of beta random variables, with the mixing distribution calculated from relative volumes of polyhedral convex cones determined by the convex shape restriction. Simulations show that the power of the test is favourable compared with the usual F‐test against a quadratic model, for some nonquadratic choices of the underlying regression function. [ABSTRACT FROM PUBLISHER]

When fitting a linear regression model to data, aliasing can adversely affect the estimates of the model coefficients and the decision of whether or not a term is significant. Optimal experimental designs give efficient estimators assuming that the true form of the model is known, while robust experimental designs guard against inaccurate estimates caused by model misspecification. Although it is rare for a single design to be both maximally efficient and robust, it is shown here that uniform designs limit the effects of aliasing to yield reasonable efficiency and robustness together. Aberration and resolution measure how well fractional factorial designs guard against the effects of aliasing. Here it is shown that the definitions of aberration and resolution may be generalised to other types of design using the discrepancy. [ABSTRACT FROM PUBLISHER]

Often in sequential trials some additional data become available after a stopping boundary has been reached. A method for incorporating such information from overrunning is developed, based on a maximum‐likelihood ordering of the sample space after overrunning. This yields a p‐value for the primary test and a median‐unbiased estimator and confidence intervals for the parameter under test. The context is that of observing a Brownian motion with drift, with either linear stopping boundaries in continuous time or discrete‐time group‐sequential boundaries. The methods apply to many clinical trials and are exemplified with data from a survival‐analysis‐based sequential clinical trial. [ABSTRACT FROM PUBLISHER]

The results of Davies (1977, 1987) are extended to a linear model situation with unknown residual variance. [ABSTRACT FROM PUBLISHER]

We propose an information‐matrix‐based goodness‐of‐fit statistic to test the validity of the logistic regression model based on case‐control data by extending the information matrix test of White (1982) for detecting one‐sample parametric model misspecification to the semiparametric profile likelihood setting under a two‐sample semiparametric model, which is equivalent to the assumed logistic regression model. The proposed test statistic requires a high‐dimensional matrix inversion, but is otherwise easily computed and has an asymptotic chi‐squared distribution. This test statistic is an alternative to the Kolmogorov–Smirnov‐type statistic of Qin & Zhang (1997) and the chi‐squared‐type statistic of Zhang (1999) and needs neither to employ a bootstrap method to evaluate its critical values nor to group the combined sample data into a finite number of mutually exclusive categories even when the underlying population distribution is continuous. We demonstrate that the proposed test statistic and its asymptotic distribution may be obtained by fitting the prospective logistic regression model to case‐control data. We present some results on simulation and on the analysis of three real datasets. [ABSTRACT FROM PUBLISHER]

For the one‐way random effects model three noniterative robust variance component estimators are investigated. For these estimators block breakdown points, measurement breakdown points and efficiency are derived. It is shown that the block breakdown points are approximating 50% and that the measurement breakdown points are not smaller than 25%. By deriving the asymptotic normal distribution of these estimators we show that two of the new estimators have an asymptotic efficiency greater than 82% and under certain circumstances up to 96%. Competing estimators have lower breakdown points and lower efficiencies. A simulation study shows that the high efficiency also appears for finite samples. The use of the estimators for interlaboratory tests is demonstrated in a practical example. [ABSTRACT FROM PUBLISHER]

In randomisation and permutation inference, pivotal arguments remove the hypothesised treatment effect, thereby basing inferences on the null distribution in which the treatment has no effect. This is common, for instance, with additive treatment effects. The current paper uses ‘attributable eects» to expand substantially the scope of pivotal arguments. Attributable effects are defined for three cases, namely the 2 × 2 contingency table, displacement eects and the Mann–Whitney–Wilcoxon statistic, and in each case removing an appropriate attributable effect restores the familiar null randomisation distribution of the associated statistic, yielding exact inferences. The procedure extends immediately for use in sensitivity analysis in nonrandomised observational studies. [ABSTRACT FROM PUBLISHER]

We propose a chi-squared-type statistic to test the validity of the logistic regression model based on case-control data by adapting the goodness-of-fit test of Nikulin-Rao-Robson-Moore. The proposed test statistic requires a high-dimensional matrix inversion, but is otherwise easy to compute and has an asymptotic chi-squared distribution. This test statistic is an alternative to the Kolmogorov-Smirnov-type statistic of Qin & Zhang (1997) and does not need to employ a bootstrap method to evaluate its critical values. We present some results on simulation and on analysis of two real datasets. [ABSTRACT FROM PUBLISHER]

Clinical trials often include interim analyses that compare treatment groups with respect to the mean function of a response process. Sometimes it is unclear how the mean functions of the groups might differ, and thus one cannot confidently prespecify a simple metric upon which a stopping rule or repeated confidence interval can be based. This motivated us to extend the repeated confidence intervals approach for a finite-dimensional parameter (Jennison & Turnbull, 1989) to the use of repeated confidence bands for the mean function of a response process. Formal tests of hypotheses are easily constructed from the repeated confidence bands. We also describe how inferences for the mean function can be adaptively restricted to different subsets of its domain at different interim analyses. An example is given involving an AIDS clinical trial. [ABSTRACT FROM PUBLISHER]

We consider construction of two-sided nonparametric confidence intervals in a smooth function model setting. A nonparametric likelihood approach based on Stein's least favourable family is proposed as an alternative to empirical likelihood. The approach enjoys the same asymptotic properties as empirical likelihood, but is analytically and computationally less cumbersome. The simplicity of the method allows us to propose and analyse asymptotic and bootstrapping techniques as a means of reducing coverage error to levels comparable with those obtained by more computationally-intensive techniques such as the iterated bootstrap. A simulation study confirms that coverage error may be substantially reduced by simple analytic adjustment of the nonparametric likelihood interval and that bootstrapping the distribution of the nonparametric likelihood ratio results in very desirable coverage accuracy. [ABSTRACT FROM PUBLISHER]

The additive risk model assumes that the hazard function associated with a set of covariates is the sum of the baseline hazard function and the regression function of covariates. We propose two different test procedures for checking the adequacy of two-sample additive risk models for randomly censored observations. One is based on the martingale residuals and the other on the difference between weighted estimators of the excess risk. The test statistics are shown to be asymptotically normal under appropriate regularity conditions and consistent under any model misspecifications. Finally, two real examples are provided, along with results of a simulation study. [ABSTRACT FROM AUTHOR]

The method of empirical likelihood can be viewed as one of allocating probabilities to an n-cell contingency table so as to minimise a goodness-of-fit criterion. It is shown that, when the Cressie-Read power-divergence statistic is used as the criterion, confidence regions enjoying the same convergence rates as those found for empirical likelihood can be obtained for the entire range of values of the Cressie-Read parameter λ, including − 1, maximum entropy, 0, empirical likelihood, and 1, Pearson's χ2. It is noted that, in the power-divergence family, empirical likelihood is the only member which is Bartlettcorrectable. However, simulation results suggest that, for the mean, using a, scaled F distribution yields more accurate coverage levels for moderate sample sizes. [ABSTRACT FROM AUTHOR]

Consider a sequential test applied to two streams of binary responses in a comparative clinical trial. After the completion of such a trial Woodroofe's (1992) technique for accurate confidence interval estimation for a treatment difference can be applied. In this paper the technique is extended to provide confidence intervals for the individual absolute success probabilities of the two treatments in the trial, for which no current methodology exists. Accuracy is explored by simulation of coverage probabilities and individual confidence limit probabilities. The simulations concern a triangular test and an O'Brien & Fleming test. [ABSTRACT FROM AUTHOR]