Projection estimators of Pickands dependence functions

The authors consider the construction of intrinsic estimators for the Pickands dependence function of an extreme‐value copula. They show how an arbitrary initial estimator can be modified to satisfy the required shape constraints. Their solution consists in projecting this estimator in the space of Pickands functions, which forms a closed and convex subset of a Hilbert space. As the solution is not explicit, they replace this functional parameter space by a sieve of finite‐dimensional subsets. They establish the asymptotic distribution of the projection estimator and its finite‐dimensional approximations, from which they conclude that the projected estimator is at least as efficient as the initial one.     

A LAN based Neyman smooth test for Pareto distributions

The Pareto distribution is found in a large number of real world situations and is also a well-known model for extreme events. In the spirit of Neyman [1937. Smooth tests for goodness of fit. Skand. Aktuarietidskr. 20, 149–199] and Thomas and Pierce [1979. Neyman's smooth goodness-of-fit test when the hypothesis is composite. J. Amer. Statist. Assoc. 74, 441–445], we propose a smooth goodness of fit test for the Pareto distribution family which is motivated by LeCam's theory of local asymptotic normality (LAN). We establish the behavior of the associated test statistic firstly under the null hypothesis that the sample follows a Pareto distribution and secondly under local alternatives using the LAN framework. Finally, simulations are provided in order to study the finite sample behavior of the test statistic.     

Bootstrap confidence intervals for the pareto index

In the present paper we develop second-order theory using the subsample bootstrap in the context of Pareto index estimation. We show that the bootstrap is not second-order accurate, in the sense that it fails to correct the first term describing departure from the limit distribution. Worse than this, even when the subsample size is chosen optimally, the error between the subsample bootstrap approximation and the true distribution is often an order of magnitude larger than that oi tue asymptotic approximation. To overcome this deficiency, we show that an extrapolation method, based quite literally on a mixture of asymptotic and subsample bootstrap methods, can lead to second-order correct confidence intervals for the Pareto index.     

Promoting social equality values to improve judgement of a woman reporting sexism / El fomento de valores relacionados con la igualdad social para mejorar la opinión sobre las mujeres que denuncian discriminación por sexismo

Previous research has shown that discriminated women blame themselves more than they blame discrimination when meritocracy values are salient. In two studies, we examined whether meritocracy values also influence female observers when they judge a female victim of sexism. Such values were expected to lead them to judge more positively a victim incriminating herself than a victim claiming discrimination. Conversely, social equality values should lead them to judge more positively a victim claiming discrimination. Women who were either feminists or non-feminists (Study 1) or who were exposed to either social equality values or personal merit values (Study 2) had to judge a female victim of sexism who ascribed what happened to discrimination or to her ability. Feminist women and women exposed to social equality judged the female victim more positively when she reported discrimination than when she incriminated herself. The reverse pattern of judgement was observed for non-feminist women and women exposed to meritocracy values. The importance of values is discussed to improve the image of women claiming sexism.     

Nonlinear mixed effects modelling for estimation of steady state attainment

Assessment of the time needed to attain steady state is a key pharmacokinetic objective during drug development. Traditional approaches for assessing steady state include ANOVA‐based methods for comparing mean plasma concentration values from each sampling day, with either a difference or equivalence test. However, hypothesis‐testing approaches are ill suited for assessment of steady state. This paper presents a nonlinear mixed effects modelling approach for estimation of steady state attainment, based on fitting a simple nonlinear mixed model to observed trough plasma concentrations. The simple nonlinear mixed model is developed and proposed for use under certain pharmacokinetic assumptions. The nonlinear mixed modelling estimation approach is described and illustrated by application to trough data from a multiple dose trial in healthy subjects. The performance of the nonlinear mixed modelling approach is compared to ANOVA‐based approaches by means of simulation techniques. Copyright © 2005 John Wiley & Sons, Ltd.     

Asymptotics of cross-validated risk estimation in estimator selection and performance assessment

Risk estimation is an important statistical question for the purposes of selecting a good estimator (i.e., model selection) and assessing its performance (i.e., estimating generalization error). This article introduces a general framework for cross-validation and derives distributional properties of cross-validated risk estimators in the context of estimator selection and performance assessment. Arbitrary classes of estimators are considered, including density estimators and predictors for both continuous and polychotomous outcomes. Results are provided for general full data loss functions (e.g., absolute and squared error, indicator, negative log density). A broad definition of cross-validation is used in order to cover leave-one-out cross-validation, V-fold cross-validation, Monte Carlo cross-validation, and bootstrap procedures. For estimator selection, finite sample risk bounds are derived and applied to establish the asymptotic optimality of cross-validation, in the sense that a selector based on a cross-validated risk estimator performs asymptotically as well as an optimal oracle selector based on the risk under the true, unknown data generating distribution. The asymptotic results are derived under the assumption that the size of the validation sets converges to infinity and hence do not cover leave-one-out cross-validation. For performance assessment, cross-validated risk estimators are shown to be consistent and asymptotically linear for the risk under the true data generating distribution and confidence intervals are derived for this unknown risk. Unlike previously published results, the theorems derived in this and our related articles apply to general data generating distributions, loss functions (i.e., parameters), estimators, and cross-validation procedures.