Crossings of a height and related bank order statistics

Former results on BAHADUR efficiency of signed rank tests are carried over to the class of two-sample rank tests. It is shown that the two-sample rank tests are asymptotically optimal at alternatives far away from the hypothesis under fairly general conditions. Surprisingly, the median test appears to be optimal only in case of equal sample sizes.     

Resampling schemes with low resampling intensity and their applications in testing hypotheses

The paper explores statistical features of different resampling schemes under low resampling intensity. The original sample is considered in a very general framework of triangular arrays, without independence or equally distributed assumptions, although improvements under such conditions are also provided. We show that low resampling schemes have very interesting and flexible properties, providing new insights into the performance of widely used resampling methods, including subsampling, two-sample unbalanced permutation statistics or wild bootstrap. It is shown that, under regularity assumptions, resampling tests with critical values derived by the appertaining low resampling procedures are asymptotically valid and there is no loss of power compared with the power function of an ideal (but unfeasible) parametric family of tests. Moreover we show that in several contexts, including regression models, they may act as a filter for the normal part of a limit distribution, turning down the influence of outliers.     

Exact Non-parametric Confidence Intervals for Quantiles with Progressive Type-II Censoring

It is shown how various exact non-parametric inferences based on order statistics in one or two random samples can be generalized to situations with progressive type-II censoring, which is a kind of evolutionary right censoring. Ordinary type-II right censoring is a special case of such progressive censoring. These inferences include confidence intervals for a given parent quantile, prediction intervals for a given order statistic of a future sample, and related two-sample inferences based on exceedance probabilities. The proposed inferences are valid for any parent distribution with continuous distribution function. The key result is that each observable uncensored order statistic that becomes available with progressive type-II censoring can be represented as a mixture with known weights of underlying ordinary order statistics. The importance of this mixture representation lies in that various properties of such observable order statistics can be deduced immediately from well-known properties of ordinary order statistics.     

A class of exceedance-type statistics for the two-sample problem

We consider here a class of test statistics based on exceeding observations and develop exceedance-type tests for the two-sample hypothesis testing problem. The exact distribution of the statistics are derived under the null hypothesis as well as under the Lehmann alternative, and then a comparative power study is carried out.     

Exact distribution-free tests for equality of several linear models

Exact tests for the equality of several linear models are developed using permutation techniques. Two cases of the linear model, characterized by either stochastic or nonstochastic predictors, are considered: the linear regression model (LRM) and the general linear model (GLM). A general class of test statistics using the volume of simplexes as the basic unit of analysis is proposed for this problem. The resulting class of statistics is shown to be a natural generalization of the multi-response permutation procedure (MRPP) test statistics which have been shown to comprise many of the statistics used in both parametric and nonparametric analysis of the standard g—sample problem. In the LRM case, exact moments of all orders are derived for the permutation distribution of any test statistic in the general class. Moment-based approximation of significance levels is shown to be computationally feasible in the simple LRM.     

Control percentile test procedures for censored data

In life testing and survival analyses which involve the use of expensive equipment the cost of continuing an experiment until all the items on test have failed can be quite high. In these situations it is reasonable to make a statistical test when a pre-specified percentile, e.g. median of the control group has been observed. This article adapts some existing procedures for complete samples to randomly censored data. The results of Lo and Singh (1985) who extended the Bahadur representation of quantiles to the censored case enable us to use the methods of Gastwirth (1968) and Hettmansperger (1973) which were based on Bahadur's result to extend the procedures of Mathisen (1943), Gart (1963) and Slivka (1970).The large sample efficiency of the control median test is the same as that of Brookmeyer and Crowley's (1982) extension of the usual median test. For the two-sample shift problem with observations following the double-exponential law, the median remains the optimum percentile to use until the censoring becomes quite heavy. On the other hand, in the two-sample scale parameter problem for data from an exponential distribution the percentile (80th in the uncensored case) yielding the asymptotically most powerful test in the family of control percentile tests no longer is optimum. The effect becomes noticeable when 25% or more of the data is censored.     

Distribution of ranks in a two-sample problem with right censored data

We consider different censoring models for a two-sample and find the joint distribution of the rank vector and number of uncensored observations under each censoring model when the distributions of life times and/or distributions of censoring of censoring variables satisfy the condition for the Lehmann type of alternatives.     

Testing nonparametric statistical functionals with applications to rank tests

The present paper discusses how nonparametric tests can be deduced from statistical functionals. Efficient and asymptotically most powerful maximin tests are derived. Their power function is calculated under implicit alternatives given by the functional for one – and two – sample testing problems. It is shown that the asymptotic power function does not depend on the special implicit direction of the alternatives but only on quantities of the functional. The present approach offers a nonparametric principle how to construct common rank tests as the Wilcoxon test, the log rank test, and the median test from special two-sample functionals. In addition it is shown that studentized permutation tests yield asymptotically valid tests for certain extended null hypotheses given by functionals which are strictly larger than the common i.i.d. null hypothesis. As example tests concerning the von Mises functional and the Wilcoxon two-sample test are treated.     

Comparison of two weibull distributions under random censoring

The two-sample problem for comparing Weibull scale parameters is studied for randomly censored data. Three different test statistics are considered and their asymptotic properties are established under a sequence of local alternatives, It is shown that both the test statistic based on the mlefs (maximum likelihood estimators) and the likelihood ratio test are asymptotically optimum. The third statistic based only on the number of failures is not, Asymptotic relative efficiency of this statistic is obtained and its numerical values are computed for uniform and Weibull censoring, Effects of uniform random censoring on the censoring level of the experiment are illus¬trated, A direct proof for the joint asymptotic normality of the mlefs of the shape and the scale parameters is also given     

Monte carlo results for conditional survival tests under randomly censored data

Within a Monte Carlo study finite sample results are obtained for different generalized rank tests based on randomly censored life time data. It is pointed out that conditional tests should be applied in practice whenever drastic differences between the censoring distributions for the underlying groups do not appear. The tests are slight modifications of known permutation tests for censored data.     

Optimal tests for carcinogenicity in a model with fatal and incidental tumours

In a recent paper Heimann and Neuhaus (Biometrika 88 (2001) 435) studied the combined test of Peto et al. (Long Term and Short Term Screening Assays for Carcinogens: A Critical Appraisal, IARC Monograph on the evaluation of the carcinogenic risk of chemicals to humans, Annex to supplement 2, WHO, Geneva, pp. 311–426) for the two sample testing problem. This test is an ad hoc test adding the log rank test statistic for the random censorship model and the Mantel Haenszel test statistic for the interval censored data model. Heimann and Neuhaus (Biometrika 88 (2001) 435) also introduced an appropriate statistical model combining randomly censored and interval censored data, to handle lethal and incidental tumours jointly.In the present paper we will derive asymptotically optimal tests for this model in a systematic way. The optimal test statistic is decomposed into two orthogonal parts, representing the fatal and incidental components. This will allow to use well known results from the literature for the fatal part, to estimate unknown quantities for the incidental part, and last but not least to compare with Peto's combined test.Moreover, we introduce conditional permutation versions of our tests which are finite sample distribution free under the null hypothesis with equal censoring and are asymptotically equivalent to their unconditional counterparts even under locally unequal censoring. Crucial for these results is a conditional limit theorem for the test statistics under local alternatives which follows from results of Strasser and Weber (Math. Methods Statist. 2 (1999) 220).     

Tests for symmetry with right censoring

Permutation tests for symmetry are suggested using data that are subject to right censoring. Such tests are directly relevant to the assumptions that underlie the generalized Wilcoxon test since the symmetric logistic distribution for log-errors has been used to motivate Wilcoxon scores in the censored accelerated failure time model. Its principal competitor is the log-rank (LGR) test motivated by an extreme value error distribution that is positively skewed. The proposed one-sided tests for symmetry against the alternative of positive skewness are directly relevant to the choice between usage of these two tests.

The permutation tests use statistics from the weighted LGR class normally used for making two-sample comparisons. From this class, the test using LGR weights (all weights equal) showed the greatest discriminatory power in simulations that compared the possibility of logistic errors versus extreme value errors.

In the test construction, a median estimate, determined by inverting the Kaplan–Meier estimator, is used to divide the data into a “control” group to its left that is compared with a “treatment” group to its right. As an unavoidable consequence of testing symmetry, data in the control group that have been censored become uninformative in performing this two-sample test. Thus, early heavy censoring of data can reduce the effective sample size of the control group and result in diminished power for discriminating symmetry in the population distribution.     

Most powerful tests for use in matched pair experiments when data may be censored

Four distribution-free tests are developed for use in matched pair experiments when data may be censored: a bootstrap based on estimates of the median difference, and three rerandomization tests. The latter include a globally almost most powerful (GAMP) test which uses the original data and two modified Gilbert-Gehan tests which use the ranks. Computation time is reduced by using a binary count to generate subsamples and by restricting subsampling to the uncensored pairs. In Monte Carlo simulations against normal alternatives, mixed normal alternatives, and exponential alternatives, the GAMP test is most powerful with light censoring, the rank test is most powerful with heavy censoring. The bootstrap degenerates to the sign test and is least powerful.     

Conditional tests in a competing risks model

Testing for equality of competing risks based on their cumulative incidence functions (CIFs) or their cause specific hazard rates (CSHRs) has been considered by many authors. The finite sample distributions of the existing test statistics are in general complicated and the use of their asymptotic distributions can lead to conservative tests. In this paper we show how to perform some of these tests using the conditional distributions of their corresponding test statistics instead (conditional on the observed data). The resulting conditional tests are initially developed for the case of k = 2 and are then extended to k > 2 by performing a sequence of two sample tests and by combining several risks into one. A simulation study to compare the powers of several tests based on their conditional and asymptotic distributions shows that using conditional tests leads to a gain in power. A real life example is also discussed to show how to implement such conditional tests.     

The Uniform-Laplace Distribution and the D-Test

A new two-sample rank test for location is proposed. This test, called the D-test, is asymptotically efficient for underlying densities which follow a “flat-topped” Laplace distribution. The D-statistic is simple to compute, and the test may be suitable when there is censoring. The D-test includes the median test as a special case.     

On the Weak Consistency of Permutation Tests

Consistency of some nonparametric tests with real variables has been studied by several authors under the assumption that population variance is finite and/or in the presence of some violations of the data exchangeability between samples. Since main inferential conclusions of permutation tests concern the actual dataset, where sample sizes are held fixed, we consider the notion of consistency in the weak version (in probability). Here, we characterize weak consistency of permutation tests assuming population mean is finite and without assuming existence of population variance. Moreover, since permutation test statistics do not require to be standardized, we do not assume that data are homoscedastic in the alternative. Several application examples to mostly used test statistics are discussed. A simulation study and some hints for robust testing procedures are also presented.     

Bootstrap and permutation rank tests for proportional hazards under right censoring

We address the testing problem of proportional hazards in the two-sample survival setting allowing right censoring, i.e., we check whether the famous Cox model is underlying. Although there are many test proposals for this problem, only a few papers suggest how to improve the performance for small sample sizes. In this paper, we do exactly this by carrying out our test as a permutation as well as a wild bootstrap test. The asymptotic properties of our test, namely asymptotic exactness under the null and consistency, can be transferred to both resampling versions. Various simulations for small sample sizes reveal an actual improvement of the empirical size and a reasonable power performance when using the resampling versions. Moreover, the resampling tests perform better than the existing tests of Gill and Schumacher and Grambsch and Therneau . The tests’ practical applicability is illustrated by discussing real data examples.

     

The split sample permutation t-tests

Without the exchangeability assumption, permutation tests for comparing two population means do not provide exact control of the probability of making a Type I error. Another drawback of permutation tests is that it cannot be used to test hypothesis about one population. In this paper, we propose a new type of permutation tests for testing the difference between two population means: the split sample permutation t-tests. We show that the split sample permutation t-tests do not require the exchangeability assumption, are asymptotically exact and can be easily extended to testing hypothesis about one population. Extensive simulations were carried out to evaluate the performance of two specific split sample permutation t-tests: the split in the middle permutation t-test and the split in the end permutation t-test. The simulation results show that the split in the middle permutation t-test has comparable performance to the permutation test if the population distributions are symmetric and satisfy the exchangeability assumption. Otherwise, the split in the end permutation t-test has significantly more accurate control of level of significance than the split in the middle permutation t-test and other existing permutation tests.     

Asymptotically optimal rank tests for the two-sample problem with randomly censored data

Using the methods of asymptotic decision theory asymptotically optimal rank tests are constructed in the two-sample testing problem for translation families and positive scale families under the assumption of equal censoring in both samples. The resulting tests have a simple form extending the known tests for un-censored data in a natural way. Relations to a recent proposal by Albers and Akritas are discussed.     

Extending permutation conditional inference to unconditional ones

In this presentation we discuss the extension of permutation conditional inferences to unconditional or population ones. Within the parametric approach this extension is possible when the data set is randomly selected by well-designed sampling procedures on well-defined population distributions, provided that their nuisance parameters have boundely complete statistics in the null hypothesis or are provided with invariant statistics. When these conditions fail, especially if selection-bias procedures are used for data collection processes, in general most of the parametric inferential extensions are wrong or misleading. We will see that, since they are provided with similarity and conditional unbiasedness properties and if correctly applicable, permutation tests may extend, at least in a weak sense, conditional to unconditional inferences.