Robust weighted one-way ANOVA: Improved approximation and efficiency

A robust test for the one-way ANOVA model under heteroscedasticity is developed in this paper. The data are assumed to be symmetrically distributed, apart from some outliers, although the assumption of normality may be violated. The test statistic to be used is a weighted sum of squares similar to the Welch [1951. On the comparison of several mean values: an alternative approach. Biometrika 38, 330-336.] test statistic, but any of a variety of robust measures of location and scale for the populations of interest may be used instead of the usual mean and standard deviation. Under the commonly occurring condition that the robust measures of location and scale are asymptotically normal, we derive approximations to the distribution of the test statistic under the null hypothesis and to its distribution under alternative hypotheses. An expression for relative efficiency is derived, thus allowing comparison of the efficiency of the test as a function of the choice of the location and scale estimators used in the test statistic. As an illustration of the theory presented here, we apply it to three commonly used robust location–scale estimator pairs: the trimmed mean with the Winsorized standard deviation; the Huber Proposal 2 estimator pair; and the Hampel robust location estimator with the median absolute deviation.     

Partially-adaptive robust estimators of location via exponential embedding

A general method is presented for constructing a location estimator which is asymptotically efficient at any two different location-scale families of symmetric distributions as well as at an appropriately defined class of distributions lying in between. The method works by embedding the two families in a comprehensive parametric model and identifying the estimator with the MLE. The case when the families are Normal and Double exponential is examined in detail.     

Asymptotic relative efficiency of the linear discriminant function under partial nonrandom classification of the training data

This paper considers the problem where the linear discriminant rule is formed from training data that are only partially classified with respect to the two groups of origin. A further complication is that the data of unknown origin do not constitute an observed random sample from a mixture of the two under- lying groups. Under the assumption of a homoscedastic normal model, the overall error rate of the sample linear discriminant rule formed by maximum likelihood from the partially classified training data is derived up to and including terms of the first order in the case of univariate feature data. This first- order expansion of the sample rule so formed is used to define its asymptotic efficiency relative to the rule formed from a completely classified random training set and also to the rule formed from a completely unclassified random set.     

An approximation for analyzing a broad class of implicitly and explicitly defined estimators

An approximation is presented that can be used to gain insight into the characteristics – such as outlier sensitivity, bias, and variability – of a wide class of estimators, including maximum likelihood and least squares. The approximation relies on a convenient form for an arbitrary order Taylor expansion in a multivariate setting. The implicit function theorem can be used to construct the expansion when the estimator is not defined in closed form. We present several finite-sample and asymptotic properties of such Taylor expansions, which are useful in characterizing the difference between the estimator and the expansion.     

On robustness and efficiency of certain statistics involving the empirical characteristic function

Summary The polar coordinates of the empirical characteristic function have been used effectively in estimation and testing. Here the efficiency and robustness properties of two simple but useful statistics involving the polar coordinates of the empirical characteristic function are examined.     

On Efficient AR Spectral Estimation for Long-Range Predictions

ABSTRACT

This article discusses the concept of asymptotic efficiency from the frequency domain point of view making use of the direct method for multiple prediction. In particular, it is shown that with the use of a new autoregressive model selection at each prediction time h, the asymptotic lower bound of the integrated relative squared error of an AR spectral estimate is attained by the order selected for multistep prediction by the AIC selection procedure Akaike (1973 Akaike , H. ( 1973 ). Information theory and an extension of the maximum likelihood principle . In Petrov , B. N. , Csaki , F. eds. Proc. 2nd Int. Symp. on Information Theory. Budapest : Akademia Kiado , 267 – 281 . [Google Scholar]) and its alike when the underlying process is a non zero mean infinite order not necessarily Gaussian AR process.     

One-sample location tests for multilevel data

In this paper, we consider testing the location parameter with multilevel (or hierarchical) data. A general family of weighted test statistics is introduced. This family includes extensions to the case of multilevel data of familiar procedures like the t, the sign and the Wilcoxon signed-rank tests. Under mild assumptions, the test statistics have a null limiting normal distribution which facilitates their use. An investigation of the relative merits of selected members of the family of tests is achieved theoretically by deriving their asymptotic relative efficiency (ARE) and empirically via a simulation study. It is shown that the performance of a test depends on the clusters configurations and on the intracluster correlations. Explicit formulas for optimal weights and a discussion of the impact of omitting a level are provided for 2 and 3-level data. It is shown that using appropriate weights can greatly improve the performance of the tests. Finally, the use of the new tests is illustrated with a real data example.     

A class of k-sample distribution-free tests for location against ordered alternatives

This paper introduces a new class of distribution-free tests for testing the homogeneity of several location parameters against ordered alternatives. The proposed class of test statistics is based on a linear combination of two-sample U-statistics based on subsample extremes. The mean and variance of the test statistic are obtained under the null hypothesis as well as under the sequence of local alternatives. The optimal weights are also determined. It is shown via Pitman ARE comparisons that the proposed class of test statistics performs better than its competitor tests in case of heavy-tailed and long-tailed distributions     

On the bias and variance of some proportion estimators

Providing certain parameters are known, almost any linear map from R^P to R¹ can be adjusted to yield a consistent and unbiased estimator in the context of estimating the mixing proportion θ on the basis of an unclassified sample of observations taken from a mixture of two p-dimensional distributions in proportions θ and 1-θ. Attention is focused on an estimator proposed recently, θ, which has minimum variance over all such linear maps. Unfortunately, the form of θ depends on the means of the component distributions and the covariance matrix of the mixture distribution. The effect of using appropriate sample estimates for these unknown parameters in forming θ is investigated by deriving the asymptotic mean and variance of the resulting estimator. The relative efficiency of this estimator under normality is derived. Also, a study is undertaken of the performance of a similar type of estimator appropriate in the context where an observed data vector is not an observation from either one or the other onent distributions, but is recorded as an integrated measurement over a surface area which is a mixture of two categories whose characteristics have different statistical distributions.The asymptotic bias in this case is compared with some available practical results.     

On the maximum likelihood estimate for the drift of brownian motion following a symmetric sequential probability ratio test

Bias and variance are evaluated explicitly for the maximum likelihood estimate (MLE) of the drift of a Brownian motion following a symmetric sequential probability ratio test (SPRT). The MLE is shown to be asymptotically efficient when the boundary of the SPRT tends to infinity.     

Influences of misspecification on asymptotic relative efficiency of coefficients of determination: Application to agriculture

Violation of correct specification may cause some undesirable results such as biased logistic regression coefficients and less efficient test statistics. In this paper, asymptotic relative efficiency (ARE) of various coefficients of determination in misspecified binary logistic regression models is investigated. Seven types of misspecification have been included. ARE of test statistics for exponential and Weibull distributions as a method of calculating optimal cutpoints is derived to demonstrate misspecification. Theoretical relationships between coefficients of determination have also been analyzed. Extensive simulations using bootstrap method and a real data application reveal more efficient one under various modeling scenarios.     

On the boundary properties of Bernstein polynomial estimators of density and distribution functions

For density and distribution functions supported on [0,1], Bernstein polynomial estimators are known to have optimal mean integrated squared error (MISE) properties under the usual smoothness conditions on the function to be estimated. These estimators are also known to be well-behaved in terms of bias: they have uniform bias over the entire unit interval. What is less known, however, is that some of these estimators do experience a boundary effect, but of a different nature than what is seen with the usual kernel estimators.     

Properties of a class of inequality measures: A proposal for a resistant version

Summary This note explores the robustness properties of a general class of ineqyality measures which includes the Bonferroni and the Gini indexes as special cases and proposes some modifications in order to make them outlier resistant.     

Relative efficiency of a censored experiment in terms of fisher information matrix

Fisher information matrix is used to quantify information loss in the randomly right-censored model, A real value approach alternative to the matrix approach of Turrero (1988) is presented for obtaining real valued measures of the relative efficiency of the censored experiment. Properties of the proposed measures are examined. The connection between both approaches and the Bayesian approach to this problem is also studied. Results in the paper are exemplified by considering grouped survival data.     

Another look at the robustness of colton's model for comparing two medical treatments

The robustness of an extended version of Colton's decision theoretic model is considered. The extended version includes the losses due to the patients who are not entered in the experiment, but require treatment while the experiment is in progress. Among the topics considered are the effects of risk of using a sample size considerably less than the optimum, use of an incorrect patient horizon, application of a modified loss function, and use of a two point prior distribution. It is shown that the investigated model is robust with respect to all these changes with the exception of the use of the modified prior density.     

Asymptotic power efficiency for a location and scale problem

The asymptotic power efficiency of the class of linear rank tests relative to the asymptotically most powerful rank test is derived for a two sample location and scale problem and numerical evaluations are presented for two special tests.     

On an adaptive transformation–retransformation estimate of multivariate location

An affine equivariant estimate of multivariate location based on an adaptive transformation and retransformation approach is studied. The work is primarily motivated by earlier work on different versions of the multivariate median and their properties. We explore an issue related to efficiency and equivariance that was originally raised by Bickel and subsequently investigated by Brown and Hettmansperger. Our estimate has better asymptotic performance than the vector of co-ordinatewise medians when the variables are substantially correlated. The finite sample performance of the estimate is investigated by using Monte Carlo simulations. Some examples are presented to demonstrate the effect of the adaptive transformation–retransformation strategy in the construction of multivariate location estimates for real data.     

Some recurrence relation-based estimators of the parameters of a generalized log-series distribution

Two families of closed form estimators are proposed for estimating the single parameter of the log-series distribution(LSD)and for estimating the two parameters of a generalization of the LSD distribution(GLSD)presented by Tripathi and Gupta(1985). These families are based on the recurrence relations obtained from these distributions, are of closed form, and have very high asymptotic relative effi¬ciencies. Some two-stage procedures are suggested.