A note on simultaneous inference with rerandomization tests

Valid simultaneous confidence intervals based on rerandomization are provided for the first time. They are derived from joint confidence regions which are constructed by testing for all possible parametric values. A simple exampe illustrates these confidence intervals and compares inferences from them with other methods.     

Conditional inference under simultaneous stochastic ordering constraints

Testing for stochastic ordering is of considerable importance when increasing does of a treatment are being compared, but in applications involving multivariate responses has received much less attention. We propose a permutation test for testing against multivariate stochastic ordering. This test is distribution-free and no assumption is made about the dependence relations among variables. A comparative simulation study shows that the proposed solution exhibits a good overall performance when compared with existing tests that can be used for the same problem.     

Large-scale multiple testing under dependence

Summary.  The paper considers the problem of multiple testing under dependence in a compound decision theoretic framework. The observed data are assumed to be generated from an underlying two-state hidden Markov model. We propose oracle and asymptotically optimal data-driven procedures that aim to minimize the false non-discovery rate FNR subject to a constraint on the false discovery rate FDR. It is shown that the performance of a multiple-testing procedure can be substantially improved by adaptively exploiting the dependence structure among hypotheses, and hence conventional FDR procedures that ignore this structural information are inefficient. Both theoretical properties and numerical performances of the procedures proposed are investigated. It is shown that the procedures proposed control FDR at the desired level, enjoy certain optimality properties and are especially powerful in identifying clustered non-null cases. The new procedure is applied to an influenza-like illness surveillance study for detecting the timing of epidemic periods.     

On non parametric statistical inference for densities under long-range dependence

Statistical inference for kernel estimators of the marginal density is considered for stationary processes with long-range dependence. The asymptotic behavior is known to differ sharply between small and large bandwidths. The statistical implications of this dichotomy have not been fully explored in the literature. The optimal rate and a functional limit theorem are obtained for large bandwidths, if the long-memory parameter exceeds a certain threshold. The threshold can be lowered arbitrarily close to the lower bound of the long-memory range. This result is extended to processes with infinite variance, and the construction of simultaneous finite-sample confidence bands is considered.     

Improved simultaneous intervals for linear combinations of parameters from generalized linear models

When employing generalized linear models, interest often focuses on estimation of odds ratios or relative risks. Additionally, researchers often make overall conclusions, requiring accurate estimation of a set of these quantities. Consequently, simultaneous estimation is warranted. Current simultaneous estimation methods only perform well in this setting when there are a very small number of comparisons and/or the sample size is relatively large. Additionally, the estimated quantities can have significant bias especially at small sample sizes. The proposed bounds: (1) perform well for a small or large number of comparisons, (2) exhibit improved performance over current methods for small to moderate sample sizes, (3) provide bias adjustment not reliant on asymptotics, and (4) avoid the infinite parameter estimates that can occur with maximum-likelihood estimators. Simulations demonstrate that the proposed bounds achieve the desired level of confidence at smaller sample sizes than previous methods.     

Simulation-based inference for simultaneous processes on regular lattices

The article proposes a simulation-based inferential method for simultaneous processes defined on a regular lattice. The focus is on spatio-temporal processes with a simultaneous component, that is such that contemporaneous spatial neighbors are potential explanatory variables in the model. The new method has the advantage of being simpler to implement than maximum likelihood and allows us to propose a robust estimator. We give asymptotic properties, present a Monte Carlo study and an illustrative example.     

Evaluating surrogate variables for improving microarray multiple testing inference

The use of surrogate variables has been proposed as a means to capture, for a given observed set of data, sources driving the dependency structure among high-dimensional sets of features and remove the effects of those sources and their potential negative impact on simultaneous inference. In this article we illustrate the potential effects of latent variables on testing dependence and the resulting impact on multiple inference, we briefly review the method of surrogate variable analysis proposed by Leek and Storey (PNAS 2008; 105:18718-18723), and assess that method via simulations intended to mimic the complexity of feature dependence observed in real-world microarray data. The method is also assessed via application to a recent Merck microarray data set. Both simulation and case study results indicate that surrogate variable analysis can offer a viable strategy for tackling the multiple testing dependence problem when the features follow a potentially complex correlation structure, yielding improvements in the variability of false positive rates and increases in power.     

Multiple robustness estimation in causal inference

Abstract

Estimation of average treatment effect is crucial in causal inference for evaluation of treatments or interventions in biostatistics, epidemiology, econometrics, sociology. However, existing estimators require either a propensity score model, an outcome vector model, or both is correctly specified, which is difficult to verify in practice. In this paper, we allow multiple models for both the propensity score models and the outcome models, and then construct a weighting estimator based on observed data by using two-sample empirical likelihood. The resulting estimator is consistent if any one of those multiple models is correctly specified, and thus provides multiple protection on consistency. Moreover, the proposed estimator can attain the semiparametric efficiency bound when one propensity score model and one outcome vector model are correctly specified, without requiring knowledge of which models are correct. Simulations are performed to evaluate the finite sample performance of the proposed estimators. As an application, we analyze the data collected from the AIDS Clinical Trials Group Protocol 175.     

Algorithms for approximate conditional inference

This paper presents a method for listing the sample space for a conditional distribution in a discrete generalized linear model. This tabulation is used in conjunction with saddlepoint methods to approximate the associated conditional probabilities. These probabilities are used to calculate conditional p-values.     

Simultaneous inference based on rank statistics in linear models

A class of simultaneous tests based on the aligned rank transform (ART) statistics is proposed for linear functions of parameters in linear models. The asymptotic distributions are derived. The stability of the finite sample behaviour of the sampling distribution of the ART technique is studied by comparing the simulated upper quantiles of its sampling distribution with those of the multivariate t-distribution. Simulation also shows that the tests based on ART have excellent small sample properties and because of their robustness perform better than the methods based on the least-squares estimates.     

Bayesian inference for multinomial populations under stochastic ordering

In many parametric problems the use of order restrictions among the parameters can lead to improved precision. Our interest is in the study of several multinomial populations under the stochastic order restriction (SOR) for univariate situations. We use Bayesian methods to show that the SOR can lead to larger gains in precision than the method without the SOR when the SOR is reasonable. Unlike frequentist order restricted inference, our methodology permits analysis even when there is uncertainty about the SOR. Our method is sampling based, and we use simple and efficient rejection sampling. The Bayes factor in favor of the SOR is computed in a simple manner, and samples from the requisite posterior distributions are easily obtained. We use real data to illustrate the procedure, and we show that there is likely to be larger gains in precision under the SOR.     

Some comments on a class of simultaneous inference procedures in ancova

Methods for constructing simultaneous confidence intervals for contrasts of treatment effects in analysis of covariance (ANCOVA) models are discussed and compared. A simple procedure is given by which, under a normality assumption made on the covariates, any method appropriate in an analysis of variance (ANOVA) may be applied to a corresponding ANCOVA by means of a simple adjustment of the required critical values. Some properties of this procedure are noted.     

Charts for finding upper percentage points of Student's t in the range .01 to .00001

The Bonferroni t-statistic is a versatile tool in multiple comparisons problems. The need for "oddball percentage points" may lead to extensive tables or heavy computation. Charts of t_p as a function of log p enable near two-decimal accuracy for any percentage point between .01 and .00001     

Exact inference for multivariate ordered alternatives

In this paper we discuss three types of ordered alternatives ordered location, stochastic ordering and quadrant dependence. We prove that quadrant dependence is the more general among the three. Then we consider a conditional tests for the equality of c distributions against quadrant dependence in a multivariate setup. An exact simultaneous testing procedure based on dependent conditional tests is presented. Two applications to real data are also given.     

Exact likelihood inference for two populations from two-parameter exponential distributions under joint Type-II censoring

In this article, we develop exact inference for two populations that have a two-parameter exponential distribution with the same location parameter and different scale parameters when Type-II censoring is implemented on the two samples in a combined manner. We obtain the conditional maximum likelihood estimators (MLEs) of the three parameters. We then derive the exact distributions of these MLEs along with their moment generating functions. Based on general entropy loss function, Bayesian study about the parameters is presented. Finally, some simulation results and an illustrative example are presented to illustrate the methods of inference developed here.     

Linear calibra and conditional inference

Approximate conditional inference is developed for the linear calibration problem. It is shown that this problem can be transformed so that the primary parameter is an angle, the nuisance parameter is a radial distance, and the density is rotationally symmetric. Were the nuisance parameter known, exact location confidence intervals would be available by location of structural arguments. A confidence distribution is used to average out the nuisance parameter yielding an approximate confidence interval that involves a precision indicator derived from the radial distance. Some difficulties with the ordinary solution are avoided by the conditional procedure.     

A Rejection Principle for Sequential Tests of Multiple Hypotheses Controlling Familywise Error Rates

We present a unifying approach to multiple testing procedures for sequential (or streaming) data by giving sufficient conditions for a sequential multiple testing procedure to control the familywise error rate (FWER). Together, we call these conditions a ‘rejection principle for sequential tests’, which we then apply to some existing sequential multiple testing procedures to give simplified understanding of their FWER control. Next, the principle is applied to derive two new sequential multiple testing procedures with provable FWER control, one for testing hypotheses in order and another for closed testing. Examples of these new procedures are given by applying them to a chromosome aberration data set and finding the maximum safe dose of a treatment.     

Robust inference from multiple test statistics via permutations: a better alternative to the single test statistic approach for randomized trials

Formal inference in randomized clinical trials is based on controlling the type I error rate associated with a single pre‐specified statistic. The deficiency of using just one method of analysis is that it depends on assumptions that may not be met. For robust inference, we propose pre‐specifying multiple test statistics and relying on the minimum p‐value for testing the null hypothesis of no treatment effect. The null hypothesis associated with the various test statistics is that the treatment groups are indistinguishable. The critical value for hypothesis testing comes from permutation distributions. Rejection of the null hypothesis when the smallest p‐value is less than the critical value controls the type I error rate at its designated value. Even if one of the candidate test statistics has low power, the adverse effect on the power of the minimum p‐value statistic is not much. Its use is illustrated with examples. We conclude that it is better to rely on the minimum p‐value rather than a single statistic particularly when that single statistic is the logrank test, because of the cost and complexity of many survival trials. Copyright © 2013 John Wiley & Sons, Ltd.     

Approximate Studentization with marginal and conditional inference

Many inference problems lead naturally to a marginal or conditional measure of departure that depends on a nuisance parameter. As a device for first-order elimination of the nuisance parameter, we suggest averaging with respect to an exact or approximate confidence distribution function. It is shown that for many standard problems where an exact answer is available by other methods, the averaging method reproduces the exact answer. Moreover, for the gamma-mean problem, where the exact answer is not explicitly available, the averaging method gives results that agree closely with those obtained from higher-order asymptotic methods. Examples are discussed; detailed asymptotic calculations will be examined elsewhere.     

Conditional inference in finite population sampling under a calibration model

Several estimators, including the classical and the regression estimators of finite population mean, are compared, both theoretically and empirically, under a calibration model, where the dependent variable(y), and not the independent variable(x), can be observed for all units of the finite population. It is shown asymptotically that when conditioned on x, the bias of the classical estimator may be much smaller than that of the regression estimators; whereas when conditioned on y, the regression estimator may have much smaller conditional bias than the classical estimator. Since all the y's(not x's) can be observed, it seems appropriate to make comparison under the conditional distribution of each estimator with y fixed. In this case, the regression estimator has smaller variance, smaller conditional bias, and the conditional coverage probability closer to its nominal level