Search

ispartof: Econometrics and Statistics status: accepted

Early Access: MAR 2021; National audience; Regression extremiles define a least squares analogue of regression quantiles. They are determined by weighted expectations rather than tail probabilities. Of special interest is their intuitive meaning in terms of expected minima and maxima. Their use appears naturally in risk management where, in contrast to quantiles, they fulfill the coherency axiom and take the severity of tail losses into account. In addition, they are comonotonically additive and belong to both the families of spectral risk measures and concave distortion risk measures. This paper provides the first detailed study exploring implications of the extremile terminology in a general setting of presence of covariates. We rely on local linear (least squares) check function minimization for estimating conditional extremiles and deriving the asymptotic normality of their estimators. We also extend extremile regression far into the tails of heavy-tailed distributions. Extrapolated estimators are constructed and their asymptotic theory is developed. Some applications to real data are provided.

When the interest is in studying conditional dependencies, and more precisely the strength of conditional dependencies, some kind of averaging over the conditioning random vector may be needed. Examples of average measures that can serve in this context are the average conditional Kendall’s tau and partial Kendall’s tau. It is known that these measures differ in general. Some statistical tests are based on these average measures, and a better knowledge of them is of importance. The aim of this paper is to provide a quantitative study of the possible differences of these two average measures, and to establish su˚cient conditions under which they coincide. Both measures are studied in two fairly general settings. In each setting theoretical results are established as well as several illustrative examples given.

A new way of constructing flexible and unimodal circular models, focusing on the modal direction, is proposed. Starting from a base symmetric density and a weight function, a two–piece four parameters density is introduced. The proposed density provides an extension of the base density to allow for sharply peaked and flat–topped unimodal distributions as well as a wide range of skewness. In particular, it generalizes some well–known peakedness–free models such as the Batschelet and Papakonstantinou densities. The four parameters of the model have a clear interpretation: modal direction, concentration, peakedness at the left and at the right of the modal direction. Symmetric submodels are obtained when the peakedness parameters are equal. The main properties related to the shape of the new density are presented and asymptotic results for maximum likelihood estimators are derived. An illustrative application concerning the flight orientation of migrating raptors is investigated.

This paper studies local polynomial estimation of expectile regression. Expectiles and quantiles both provide a full characterization of a (conditional) distribution function, but have each their own merits and inconveniences. Local polynomial fitting as a smoothing technique has a major advantage of being simple, allowing for explicit expressions and henceforth advantages when doing inference theory. The aim of this paper is twofold: to study in detail the use of local polynomial fitting in the context of expectile regression and to contribute to the important issue of bandwidth selection, from theoretical and practical points of view. We discuss local polynomial expectile regression estimators and establish an asymptotic normality result for them. The finite-sample performance of the estimators, combined with various bandwidth selectors, is investigated in a simulation study. Some illustrations with real data examples are given.

ispartof: ANNALS OF STATISTICS vol:50 issue:2 pages:754-780 status: published

This article studies M-type estimators for fitting robust additive models in the presence of anomalous data. The components in the additive model are allowed to have different degrees of smoothness. We introduce a new class of wavelet-based robust M-type estimators for performing simultaneous additive component estimation and variable selection in such inhomogeneous additive models. Each additive component is approximated by a truncated series expansion of wavelet bases, making it feasible to apply the method to nonequispaced data and sample sizes that are not necessarily a power of 2. Sparsity of the additive components together with sparsity of the wavelet coefficients within each component (group), results into a bi-level group variable selection problem. In this framework, we discuss robust estimation and variable selection. A two-stage computational algorithm, consisting of a fast accelerated proximal gradient algorithm of coordinate descend type, and thresholding, is proposed. When using nonconvex redescending loss functions, and appropriate nonconvex penalty functions at the group level, we establish optimal convergence rates of the estimates. We prove variable selection consistency under a weak compatibility condition for sparse additive models. The theoretical results are complemented with some simulations and real data analysis, as well as a comparison to other existing methods.

International audience; In functional data the interest is to find a global mean pattern, but also to capture the individual curve differences in phase and amplitude. This can be done conveniently by building in random effects on two levels: in the warping functions to account for individual phase variations; and in the linear structure to deal with individual amplitude variations. Via an appropriate choice of the warping function and B-spline approximations, estimation in the nonlinear mixed effects functional model is feasible, and does not require any prior knowledge on landmarks for the functional data. Sufficient and necessary conditions for identifiability of the flexible model are provided. A theoretical study is conducted: we establish asymptotic normality and consistency of the estimators of the registration and amplitude models, convergence of the iterative process, and consistency of the final estimator provided by the iterative process. The finite-sample performance of the proposed estimation procedure is investigated in a simulation study, which includes comparisons with existing methods. The added value of the developed method is further illustrated via the analysis of a real data example.

We introduce a new class of robust M-estimators for performing simultaneous parameter estimation and variable selection in high-dimensional regression models. We first explain the motivations for the key ingredient of our procedures which are inspired by regularization methods used in wavelet thresholding in noisy signal processing. The derived penalized estimation procedures are shown to enjoy theoretically the oracle property both in the classical finite dimensional case as well as the high-dimensional case when the number of variables p is not fixed but can grow with the sample size n, and to achieve optimal asymptotic rates of convergence. A fast accelerated proximal gradient algorithm, of coordinate descent type, is proposed and implemented for computing the estimates and appears to be surprisingly efficient in solving the corresponding regularization problems including the case for ultra high-dimensional data where $$p \gg n$$ . Finally, a very extensive simulation study and some real data analysis, compare several recent existing M-estimation procedures with the ones proposed in the paper, and demonstrate their utility and their advantages.

We investigate the inference of the copula parameter in the semiparametric Gaussian copula model when the copula component, subject to the influence of a covariate, is only indirectly observed as the response in a linear regression model. We consider estimators based on residual ranks instead of the usual but unobservable oracle ranks. We first study two such estimators for the copula correlation matrix, one via inversion of Spearman’s rho and the other via normal scores rank correlation estimator. We show that these estimators are asymptotically equivalent to their counterparts based on the oracle ranks. Then, for the copula correlation matrix under constrained parametrizations, we show that the classical one-step estimator in conjunction with the residual ranks remains semiparametrically efficient for estimating the copula parameter. The accuracy of the estimators based on residual ranks is confirmed by simulation studies.

Allowing for symmetry in distributions is often a necessity in statistical modelling. This paper studies a broad family of asymmetric densities, which in a regression setting shares basic philosophy with generalized (non)linear models. The main focus, however, for the family of densities studied here is quantile estimation instead of mean estimation. In a similar fashion, a broad family of conditional densities is considered in the regression setting. We discuss estimation of the parameters in the unconditional case, and establish an asymptotic normality result, with explicit expression for the asymptotic variance–covariance matrix. In the regression setting, we allow for flexible modelling and estimate nonparametrically the location and scale functions, leading to semiparametric estimation of conditional quantiles, again in the unifying framework of the considered broad family. The practical use of the proposed methods is illustrated in a real data application on locomotor performance in small and large terrestrial mammals.

© 2018 Oxford University Press. All rights reserved. Automated or data-driven bandwidth selection methods tend to break down in the presence of correlated errors. While this problem has previously been studied in the fixed design setting for kernel regression, the results were applicable only when there is knowledge about the correlation structure. This article generalizes these results to the random design setting and addresses the problem in situations where no prior knowledge about the correlation structure is available.We establish the asymptotic optimality of our proposed bandwidth selection criterion based on kernels K satisfying K(0) = 0. ispartof: Biometrika vol:105 issue:3 pages:681-690 status: published

© 2017 The Korean Statistical Society A strong law of large numbers for continuous random functions, and associated tensor product surfaces is established in the setup of discretely observed functional data. The result is shown in the framework of uniform convergence of functions, and stated without imposing any distributional assumptions. It is demonstrated that, under mild conditions, laws of large numbers for continuously observed functional data imply the corresponding laws under the discrete observational design of functions. Applications to the problem of estimation of expectation functions and covariance surfaces for discretely observed functional data are discussed. ispartof: Journal of the Korean Statistical Society vol:46 issue:4 pages:562-572 status: published

© 2017 Statistical Society of Canada In mean regression the characteristic of interest is the conditional mean of the response given the covariates. In quantile regression the aim is to estimate any quantile of the conditional distribution function. For given covariates, the conditional quantile function fully characterizes the entire conditional distribution function, in contrast to the mean which is just one of its characteristic quantities. Regression quantiles substantially out-perform the least-squares estimator for a wide class of non-Gaussian error distributions. In this article we consider quantile varying coefficient models (VCMs) that are an extension of classical quantile linear regression models, in which one allows the coefficients to depend on other variables. We consider VCMs with various structures for the variance of the errors (the variability function) in order to allow for heteroscedasticity. For longitudinal data, the time (T) dependent coefficient functions in the signal and the variability functions are estimated with P-splines (Penalized B-splines). Consistency of the proposed estimators is proved. Further, likelihood-ratio-type tests are considered for comparing the variability functions. The performance of the testing procedure is illustrated on simulated and real data. The Canadian Journal of Statistics 46: 246–264; 2018 © 2017 Statistical Society of Canada. ispartof: The Canadian Journal of Statistics vol:46 issue:2 pages:246-264 status: published

Quantile regression is an important tool in data analysis. Linear regression, or more generally, parametric quantile regression imposes often too restrictive assumptions. Nonparametric regression avoids making distributional assumptions, but might have the disadvantage of not exploiting distributional modelling elements that might be brought in. A semiparametric approach towards estimating conditional quantile curves is proposed. It is based on a recently studied large family of asymmetric densities of which the location parameter is a quantile (and not a mean). Passing to conditional densities and exploiting local likelihood techniques in a multiparameter functional setting then leads to a semiparametric estimation procedure. For the local maximum likelihood estimators the asymptotic distributional properties are established, and it is discussed how to assess finite sample bias and variance. Due to the appealing semiparametric framework, one can discuss in detail the bandwidth selection issue, and provide several practical bandwidth selectors. The practical use of the semiparametric method is illustrated in the analysis of maximum winds speeds of hurricanes in the North Atlantic region, and of bone density data. A simulation study includes a comparison with nonparametric local linear quantile regression as well as an investigation of robustness against miss-specifying the parametric model part.

© 2017 American Statistical Association, Institute of Mathematical Statistics, and Interface Foundation of North America. A major drawback of many established depth functionals is their ineffectiveness in identifying functions outlying merely in shape. Herein, a simple modification of functional depth is proposed to provide a remedy for this difficulty. The modification is versatile, widely applicable, and introduced without imposing any assumptions on the data, such as differentiability. It is shown that many favorable attributes of the original depths for functions, including consistency properties, remain preserved for the modified depths. The powerfulness of the new approach is demonstrated on a number of examples for which the known depths fail to identify the outlying functions. Supplementary material for this article is available online. ispartof: Journal of Computational and Graphical Statistics vol:26 issue:4 pages:883-893 status: published

This paper concerns a robust variable selection method in multiple linear regression: the robust S-nonnegative garrote variable selection method. In this paper the consistency of the method, both in terms of estimation and in terms of variable selection, is established. Moreover, the robustness properties of the method are further investigated by providing a lower bound for the breakdown point, and by deriving the influence function. The provided expressions nicely reveal the impact that the choice of an initial estimator has on the robustness properties of the variable selection method. Illustrative examples of influence functions for the S-nonnegative garrote as well as for the original (non-robust) nonnegative garrote variable selection method are provided. This research was supported by the IAP Research Network IAP P7/06 of the Belgian State (Belgian Science Policy), project GOA/12/014 of the KU Leuven Research Fund, and grant no. Q2 1.5.137.13N of the Flemish Research Foundation (FWO).

The interest is in regression quantiles in varying coefficient models for analyzing longitudinal data. The coefficients are allowed to vary with time, and the error variance (the variability function) varies with the covariates to allow for heteroscedasticity. The functional coefficients are estimated using penalized splines (P-splines), not requiring specification of the error distribution. A likelihood-ratio-type test is considered to test the shape (constancy, monotonicity and/or convexity) of the functional coefficients. Further, testing procedures based on L1-norm, L2-norm and L∞-norm of the differences of the P-splines coefficients are considered to test for constant functional coefficients. These norm based tests perform better than the likelihood-ratio-type test in our simulation study. An extreme value test for testing monotonicity or convexity, also performs better than the likelihood-ratio-type test. The likelihood-ratio-type test is, however, useful when testing the shape of the coeffi- cients in signal and in variability function simultaneously. A real data example demonstrates the testing procedures. The authors gratefully acknowledge support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy). I. Gijbels acknowledges support from the KU Leuven Research Council (GOA/12/014). A. Verhasselt acknowledges support from the The authors gratefully acknowledge support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy). I. Gijbels acknowledges support from the KU Leuven Research Council (GOA/12/014). A. Verhasselt acknowledges support from the Fonds Wetenschappelijk Onderzoek (FWO) research grant 1518917N. M.A. Ibrahim and A. Verhasselt acknowledge support from the Special Research Fund (BOF) of Hasselt University [14NO6VHAA]. The infrastructure of the VSC-Flemish Supercomputer Center, funded by the Hercules Foundation and the Flemish Government - department EWI, was used for the simulations.

© 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group. The paper concerns robust estimation and variable selection in heteroscedastic linear regression models. After a brief review of existing methods for estimation in such models, a robust S-estimation approach is discussed. For all methods concise descriptions of algorithms are provided. Little is available upon robust variable selection methods for heteroscedastic linear models. The paper gives essential contributions in the area of simultaneous robust estimation and variable selection, relying on basics of the nonnegative garrote method which has been proven to have very good practical as well as theoretical properties in the homoscedastic linear model context. Several numerical examples, simulations and analysis of real data, demonstrate the performances and practical use of the discussed methods. Moreover, we provide expressions for the influence functions of the estimators of the mean and the error variance parameters. Influence functions are plotted in a simple setting providing insights in the sensitivity of the estimators for a single outlying observation. ispartof: STATISTICS vol:53 issue:3 pages:489-532 status: published

© 2018, © 2018 American Statistical Association. Quantiles and expectiles of a distribution are found to be useful descriptors of its tail in the same way as the median and mean are related to its central behavior. This article considers a valuable alternative class to expectiles, called extremiles, which parallels the class of quantiles and includes the family of expected minima and expected maxima. The new class is motivated via several angles, which reveals its specific merits and strengths. Extremiles suggest better capability of fitting both location and spread in data points and provide an appropriate theory that better displays the interesting features of long-tailed distributions. We discuss their estimation in the range of the data and beyond the sample maximum. A number of motivating examples are given to illustrate the utility of estimated extremiles in modeling noncentral behavior. There is in particular an interesting connection with coherent measures of risk protection. Supplementary materials for this article are available online. ispartof: Journal of the American Statistical Association vol:114 issue:527 pages:1366-1381 status: published

In this paper, we provide a detailed study of a general family of asymmetric densities. In the general framework, we establish expressions for important characteristics of the distributions and discuss estimation of the parameters via method‐of‐moments as well as maximum likelihood estimation. Asymptotic normality results for the estimators are provided. The results under the general framework are then applied to some specific examples of asymmetric densities. The use of the asymmetric densities is illustrated in a real‐data analysis. This research was supported by the FWO (Fonds Wetenschappelijk Onderzoek) research project G.0826.15N (Flemish Science Foundation) and the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy; Federaal Wetenschapsbeleid). The first author gratefully acknowledges support from the GOA/12/014 project of the Research Fund KU Leuven (Onderzoeksraad, KU Leuven). The third author acknowledges support from the Flemish Science Foundation (FWO research grant 1518917N) and from the Special Research Fund (Bijzonder Onderzoeksfonds) of Hasselt University.

In this paper the interest is to elaborate on the generalization of bivariate association measures, namely Spearman’s rho, Kendall’s tau, Blomqvist’s beta and Gini’s gamma, for a general dimension d ≥ 2 . Desirable properties and axioms for such generalizations are discussed, where special attention is given to the impact of the addition of: (i) an independent random variable to a random vector; (ii) a conical combination of all components; (iii) a set of arbitrary random components. Existing generalizations are evaluated with respect to the axiom set. For a d -variate Gini’s gamma, a simplified formula is developed, making its analytical computation easier. Further, for Archimedean and meta-elliptical copulas the asymptotic behaviour when the dimension d increases is studied. Nonparametric estimation of the considered generalizations of multivariate association measures is reviewed and a nonparametric estimator of the multivariate Gini’s gamma is introduced. The practical use of multivariate association measures is illustrated on a real data example.

© 2016 Taylor & Francis. The smoothness of Tukey depth contours is a regularity condition often encountered in asymptotic theory, among others. This condition ensures that the Tukey depth fully characterizes the underlying multivariate probability distribution. In this paper we demonstrate that this regularity condition is rarely satisfied. It is shown that even well-behaved probability distributions with symmetrical, smooth and (strictly) quasi-concave densities may have non-smooth Tukey depth contours, and that the smoothness behaviour of depth contours is fairly unpredictable. peerreview_statement: The publishing and review policy for this title is described in its Aims & Scope. aims_and_scope_url: http://www.tandfonline.com/action/journalInformation?show=aimsScope&journalCode=gsta20 ispartof: Statistics vol:50 issue:5 pages:1075-1085 status: published

Multivariate tail coefficients are an important tool when investigating dependencies between extreme events for different components of a random vector. Although bivariate tail coefficients are well-studied, this is, to a lesser extent, the case for multivariate tail coefficients. This paper contributes to this research area by (i) providing a thorough study of properties of existing multivariate tail coefficients in the light of a set of desirable properties, (ii) proposing some new multivariate tail measurements, (iii) dealing with estimation of the discussed coefficients and establishing asymptotic consistency, and, (iv) studying the behavior of tail measurements with increasing dimension of the random vector. A set of illustrative examples is given, and practical use of the tail measurements is demonstrated in a data analysis with a focus on dependencies between stocks that are part of the EURO STOXX 50 market index.

© 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group. We provide a computational framework for the selection of weights (W1, . . .,Wd) that minimize the expected shortfall of the aggregated risk Z= ∑i=id WiXi. Contrary to classic and recent results, we neither restrict the marginal distributions nor the dependence structure of (X1, . . .,Xd) to any specific type. While the margins can be set to any absolutely continuous random variable with finite expectation, the dependence structure can be modelled by any absolutely continuous copula function. A real-world application to portfolio selection illustrates the usability of the new framework. ispartof: Applied Mathematical Finance vol:25 issue:1 pages:66-106 status: published

Quantile regression is an important tool for describing the characteristics of conditional distributions. Population conditional quantile functions cannot cross for different quantile orders. Unfortunately estimated regression quantile curves often violate this and cross each other, which can be very annoying for interpretations and further analysis. In this paper we are concerned with flexible varying-coefficient modelling, and develop methods for quantile regression that ensure that the estimated quantile curves do not cross. A second aim of the paper is to allow for some heteroscedasticity in the error modelling, and to also estimate the associated variability function. We investigate the finite-sample performances of the discussed methods via simulation studies. Some applications to real data illustrate the use of the methods in practical settings. This research is supported by the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy), and project GOA/12/014 of the Research Fund of the KU Leuven. The authors thank the Editor, an Associate Editor and the anonymous reviewers for their very valuable comments which led to a considerable improvement of the paper.

© Institute of Mathematical Statistics, 2017. In this paper, we provide an elaboration on the desirable properties of statistical depths for functional data. Although a formal definition has been put forward in the literature, there are still several unclarities to be tackled, and further insights to be gained. Herein, a few interesting connections between the wanted properties are found. In particular, it is demonstrated that the conditions needed for some desirable properties to hold are extremely demanding, and virtually impossible to be met for common depths. We establish adaptations of these properties which prove to be still sensible, and more easily met by common functional depths. ispartof: Statistical Science vol:32 issue:4 pages:630-639 status: published

© 2015 Elsevier Inc. In the analysis of functional data, the concept of data depth is of importance. Strong consistency of a sample version of a data depth is among the basic statistical properties that need to hold. In this paper we discuss consistency properties of three popular types of functional depth: the band depth, the half-region depth and the infimal depth. The latter is a special case of the recently introduced general class of Φ-depths. All three considered depth functions are of a non-integrated type. Counterexamples illustrate some problems with consistency results for these data depths. The main contribution of this paper consists of providing sufficient conditions for consistency of these non-integrated data depths to hold. publisher: Elsevier articletitle: Consistency of non-integrated depths for functional data journaltitle: Journal of Multivariate Analysis articlelink: http://dx.doi.org/10.1016/j.jmva.2015.05.012 content_type: article copyright: Copyright © 2015 Elsevier Inc. All rights reserved. ispartof: Journal of Multivariate Analysis vol:140 pages:259-282 status: published

© 2014 Elsevier B.V. All rights reserved. Robust selection of variables in a linear regression model is investigated. Many variable selection methods are available, but very few methods are designed to avoid sensitivity to vertical outliers as well as to leverage points. The nonnegative garrote method is a powerful variable selection method, developed originally for linear regression but recently successfully extended to more complex regression models. The method has good performances and its theoretical properties have been established. The aim is to robustify the nonnegative garrote method for linear regression as to make it robust to vertical outliers and leverage points. Several approaches are discussed, and recommendations towards a final good performing robust nonnegative garrote method are given. The proposed method is evaluated via a simulation study that also includes a comparison with existing methods. The method performs very well, and often outperforms existing methods. A real data application illustrates the use of the method in practice. publisher: Elsevier articletitle: Robust nonnegative garrote variable selection in linear regression journaltitle: Computational Statistics & Data Analysis articlelink: http://dx.doi.org/10.1016/j.csda.2014.11.009 content_type: article copyright: Copyright © 2014 Elsevier B.V. All rights reserved. ispartof: Computational Statistics & Data Analysis vol:85 pages:1-22 status: published

© 2015, Springer-Verlag Berlin Heidelberg. This paper concerns depth functions suitable for smooth functional data. We suggest a modification of the integrated data depth that takes into account the shape properties of the functions. This is achieved by including a derivative(s) into the definition of the suggested depth measures. We then further investigate the use of integrated data depths in supervised classification problems. The performances of classification rules based on different data depths are investigated, both in simulated and real data sets. As the proposed depth function provides a natural alternative to the depth function based on random projections, the difference in the performances of these two methods are discussed in more detail. ispartof: Computational Statistics vol:30 issue:4 pages:1011-1031 status: published

© 2016, Springer-Verlag Berlin Heidelberg. Varying coefficient models are flexible models to describe the dynamic structure in longitudinal data. Quantile regression, more than mean regression, gives partial information on the conditional distribution of the response given the covariates. In the literature, the focus has been so far mostly on homoscedastic quantile regression models, whereas there is an interest in looking into heteroscedastic modelling. This paper contributes to the area by modelling the heteroscedastic structure and estimating it from the data, together with estimating the quantile functions. The use of the proposed methods is illustrated on real-data applications. The finite-sample behaviour of the methods is investigated via a simulation study, which includes a comparison with an existing method. ispartof: Advances in Statistical Analysis vol:101 issue:2 pages:151-176 status: published

© 2016 Informa UK Limited, trading as Taylor & Francis Group. In dependence modelling using conditional copulas, one often imposes the working assumption that the covariate influences the conditional copula solely through the marginal distributions. This so-called (pairwise) simplifying assumption is almost standardly made in vine copula constructions. However, in recent literature evidence was provided that such an assumption might not be justified. Among the first issues is thus to test for its appropriateness. In this paper nonparametric tests for the null hypothesis of the simplifying assumption are proposed, and their asymptotic behaviours, under the null hypothesis and under some local alternatives, are established. The tests are fully nonparametric in nature: not requiring choices of copula families nor knowledge of the marginals. In a simulation study, the finite-sample size and power performances of the tests are investigated, and compared with these of the few available tests. A real data application illustrates the use of the tests. peerreview_statement: The publishing and review policy for this title is described in its Aims & Scope. aims_and_scope_url: http://www.tandfonline.com/action/journalInformation?show=aimsScope&journalCode=gsta20 ispartof: Statistics vol:51 issue:3 pages:475-509 status: published

We consider varying coefficient models, which are an extension of the classical linear regression models in the sense that the regression coefficients are replaced by functions in certain variables (often time). Varying coefficient models have been popular in longitudinal data and panel data studies, and have been applied in fields such as finance and health sciences. We estimate the coefficient functions by splines. An important question in a varying coefficient model is whether a coefficient function is monotone or convex. We develop consistent testing procedures for monotonicity and convexity. Moreover, we provide procedures to test simultaneously the shapes of certain coefficient functions in a varying coefficient model. The tests use constrained and unconstrained regression splines. The performances of the proposed tests are illustrated on simulated data. We also give a real data application. We would like to thank the Editor, an Associate Editor, and the referees for their detailed reading and very valuable comments on the manuscript. M. Ahkim's research was supported by the Special Research Fund (BOF) of Universiteit Antwerpen (Grant Number 42FA070300FFB5994). A. Verhasselt and I. Gijbels gratefully acknowledge support from the IAP Research Network P7/06 of the Belgian State (Belgian Science Policy). I. Gijbels acknowledges support from the KU Leuven Research Council (GOA/12/014). The infrastructure of the VSC-Flemish Supercomputer Center, funded by the Hercules Foundation and the Flemish Government-department EWI, was used for the simulations.

© 2017 Elsevier Inc. We consider copula modeling of the dependence between two or more random variables in the presence of a multivariate covariate. The dependence parameter of the conditional copula possibly depends on the value of the covariate vector. In this paper we develop a new testing methodology for some important parametric specifications of this dependence parameter: constant, linear, quadratic, etc. in the covariate values, possibly after transformation with a link function. The margins are left unspecified. Our novel methodology opens plenty of new possibilities for testing how the conditional copula depends on the multivariate covariate and also for variable selection in copula model building. The suggested test is based on a Rao-type score statistic and regularity conditions are given under which the test has a limiting chi-square distribution under the null hypothesis. For small and moderate sample sizes, a permutation procedure is suggested to assess significance. In simulations it is shown that the test performs well (even under misspecification of the copula family and/or the dependence parameter structure) in comparison to available tests designed for testing for constancy of the dependence parameter. The test is illustrated on a real data set on concentrations of chemicals in water samples. ispartof: Journal of Multivariate Analysis vol:159 pages:111-133 status: published

© 2016 EDP Sciences, SMAI. Several depths suitable for infinite-dimensional functional data that are available in the literature are of the form of an integral of a finite-dimensional depth function. These functionals are characterized by projecting functions into low-dimensional spaces, taking finite-dimensional depths of the projected quantities, and finally integrating these projected marginal depths over a preset collection of projections. In this paper, a general class of integrated depths for functions is considered. Several depths for functional data proposed in the literature during the last decades are members of this general class. A comprehensive study of its most important theoretical properties, including measurability and consistency, is given. It is shown that many, but not all, properties of the integrated depth are shared with the finite-dimensional depth that constitutes its building block. Some pending measurability issues connected with all integrated depth functionals are resolved, a broad new notion of symmetry for functional data is proposed, and difficulties with respect to consistency results are identified. A general universal consistency result for the sample depth version, and for the generalized median, for integrated depth for functions is derived. ispartof: ESAIM-Probability and Statistics vol:20 pages:95-130 status: published

© 2015 Elsevier Inc. A general result on weak convergence of the empirical measure of discretely observed functional data is shown. It is applied to the problem of estimation of functional mean value, and the problem of consistency of various types of depth for functional data. Counterexamples illustrating the fact that the assumptions as stated cannot be dropped easily are given. publisher: Elsevier articletitle: Weak convergence of discretely observed functional data with applications journaltitle: Journal of Multivariate Analysis articlelink: http://dx.doi.org/10.1016/j.jmva.2015.06.006 content_type: article copyright: Copyright © 2015 Elsevier Inc. All rights reserved. ispartof: Journal of Multivariate Analysis vol:146 pages:46-62 status: published

© 2016 Institute of Mathematical Statistics. All rights reserved. When describing adequately complex data structures one is often confronted with the fact that mean as well as variance (or more generally dispersion) is highly influenced by some covariates. Drawbacks of the available methods is that they are often based on approximations and hence a theoretical study should deal with also studying these approximations. This however is often ignored, making the statistical inference incomplete. In the proposed framework of double generalized modelling based on proper dispersion models we avoid this drawback and as such are in a good position to use recent results on Bregman divergence for establishing theoretical results for the proposed estimators in fairly general settings. We also study variable selection when there is a large number of covariates, with this number possibly tending to infinity with the sample size. The proposed estimation and selection procedure is investigated via a simulation study, that includes also a comparative study with competitors. The use of the methods is illustrated via some real data applications. ispartof: Electronic Journal of Statistics vol:10 issue:1 pages:1630-1676 status: published

Positive quadrant dependence is a specific dependence structure that is of practical importance in for example modelling dependencies in insurance and actuarial sciences. This dependence structure imposes a constraint on the copula function. The interest in this paper is to test for positive quadrant dependence. One way to assess the distribution of the test statistics under the null hypothesis of positive quadrant dependence is to resample from a constrained copula. This requires constrained estimation of a copula function. We show that this use of resampling under a constrained copula improves considerably the power performance of existing testing procedures. We propose two resampling procedures, one based on a parametric constrained copula estimation and one relying on nonparametric estimation of a positive quadrant dependence copula, and discuss their properties. The finite-sample performances of the resulting testing procedures are evaluated via a simulation study that also includes comparisons with existing tests. Finally, a data set of Danish fire insurance claims is tested for positive quadrant dependence. The Canadian Journal of Statistics 41: 36–64; 2013 © 2012 Statistical Society of Canada La dependance quadratique positive est une structure specifique de dependance, d'importance pratique dans, par example, la modelisation de dependance dans les sciences d'assurances. Une telle structure de dependance impose une contrainte sur la fonction copule. L'interet dans ce papier est de tester l'hypothese de dependance quadratique positive. Une facon pour acceder aux distributions des statistiques de test sous l'hypothese nulle de dependance quadratique positive est de faire un reechantillonage de la copula restrainte. Ceci necissite l'estimation sous contrainte de la fonction copule. Nous montrons que l'utilisation du reechantillonage a partir d'une copula contrainte ameliore considerablement les performances de puissance des procedures de test existantes. Nous proposons deux procedures de reechantillonage, une basee sur l'estimation restrainte parametrique d'une copule, et une basee sur une estimation nonparametrique d'une copule du type dependance quadratique positive, et nous discutons de leurs proprietes. Les performances de fini-sample des procedures de test sont evaluer via une etude de simulation qui inclut des comparaisons avec des tests existants. Finalement, un ensemble de donnees de reclamations d'assurance dans le cas d'incendis en Danmark, est tester pour des structures de dependance du type quadratique positive. La revue canadienne de statistique 41: 36–64; 2013 © 2012 Societe statistique du Canada

The manner in which two random variables influence one another often depends on covariates. A way to model this dependence is via a conditional copula function. This paper contributes to the study of semiparametric estimation of conditional copulas by starting from a parametric copula function in which the parameter varies with a covariate, and leaving the marginals unspecified. Consequently, the unknown parts in the model are the parameter function and the unknown marginals. The authors use a local pseudo-likelihood with nonparametrically estimated marginals approximating the unknown parameter function locally by a polynomial. Under this general setting, they prove the consistency of the estimators of the parameter function as well as its derivatives; they also establish asymptotic normality. Furthermore, they derive an expression for the theoretical optimal bandwidth and discuss practical bandwidth selection. They illustrate the performance of the estimation procedure with data-driven bandwidth selection via a simulation study and a real-data case.

Many univariate robust estimators are based on quantiles. As already theoretically pointed out by Fernholz (in J. Stat. Plan. Inference 57(1), 29-38, 1997), smoothing the empirical distribution function with an appropriate kernel and bandwidth can reduce the variance and mean squared error (MSE) of some quantile-based estimators in small data sets. In this paper we apply this idea on several robust estimators of location, scale and skewness. We propose a robust bandwidth selection and bias reduction procedure. We show that the use of this smoothing method indeed leads to smaller MSEs, also at contaminated data sets. In particular, we obtain better performances for the medcouple which is a robust measure of skewness that can be used for outlier detection in skewed distributions. © 2012 Sociedad de Estadística e Investigación Operativa. ispartof: Test vol:22 issue:3 pages:448-465 status: published

We analyse data on abortion rate (AR) in Italy with a particular focus on different behaviours in different regions in Italy. The aim is to try to reveal the relationship between the AR and several covariates that describe in some way the modernity of the region and the condition of the women there. The data are mostly underdispersed and the degree of underdispersion also varies with the covariates. To analyse these data, recent techniques for flexible modelling of a mean and dispersion function in a double exponential family framework are further developed now in a generalized additive model context for dealing with the multivariate set-up. The appealing unified framework and approach even allow to semi-parametric modelling of the covariates without any additional efforts. The methodology is illustrated on ozone-level data and leads to interesting findings in the Italian abortion data.

Abstarct. This paper is concerned with studying the dependence structure between two random variables Y1 and Y2 conditionally upon a covariate X. The dependence structure is modelled via a copula function, which depends on the given value of the covariate in a general way. Gijbels et al. (Comput. Statist. Data Anal., 55, 2011, 1919) suggested two non-parametric estimators of the ‘conditional’ copula and investigated their numerical performances. In this paper we establish the asymptotic properties of the proposed estimators as well as conditional association measures derived from them. Practical recommendations for their use are then discussed.

Generalized Linear Models are a widely used method to obtain parametric es- timates for the mean function. They have been further extended to allow the re- lationship between the mean function and the covariates to be more flexible via Generalized Additive Models. However the fixed variance structure can in many cases be too restrictive. The Extended Quasi-Likelihood (EQL) framework allows for estimation of both the mean and the dispersion/variance as functions of covari- ates. As for other maximum likelihood methods though, EQL estimates are not resistant to outliers: we need methods to obtain robust estimates for both the mean and the dispersion function. In this paper we obtain functional estimates for the mean and the dispersion that are both robust and smooth. The performance of the proposed method is illustrated via a simulation study and some real data examples.

One way to model a dependence structure is through the copula function which is a mean to capture the dependence structure in the joint distribution of variables. Association measures such as Kendall's tau or Spearman's rho can be expressed as functionals of the copula. The dependence structure between two variables can be highly influenced by a covariate, and it is of real interest to know how this dependence structure changes with the value taken by the covariate. This motivates the need for introducing conditional copulas, and the associated conditional Kendall's tau and Spearman's rho association measures. After the introduction and motivation of these concepts, two nonparametric estimators for a conditional copula are proposed and discussed. Then nonparametric estimates for the conditional association measures are derived. A key issue is that these measures are now looked at as functions in the covariate. The performances of all estimators are investigated via a simulation study which also includes a data-driven algorithm for choosing the smoothing parameters. The usefulness of the methods is illustrated on two real data examples.

In this paper the interest is in testing the null hypothesis of positive quadrant dependence between two random variables. Such a testing problem is important since prior knowledge of positive quadrant dependence is a qualitative restriction that should be taken into account in further statistical analysis, e.g. when choosing an appropriate copula function to model the dependence structure. The key methodology of the proposed testing procedures consists of evaluating a 'distance' between a nonparametric estimator of a copula and the independence copula, which serves as a reference case in the whole set of copulas having the positive quadrant dependence property. Choices of appropriate distances and nonparametric estimators of copula are discussed, and the proposed methods are compared with testing procedures based on bootstrap and multiplier techniques. The consistency of the testing procedures is established. In a simulation study the authors investigate the finite sample size and power performances of three types of test statistics, Kolmogorov-Smirnov, Cramer-von-Mises and Anderson-Darling statistics, together with several nonparametric estimators of a copula, including recently developed kernel type estimators. Finally they apply the testing procedures on some real data.

For nonparametric estimation of a smooth regression function, local linear fitting is a widely-used method. The goal of this paper is to briefly review how to use this method when the unknown curve possibly has some irregularities, such as jumps or peaks, at unknown locations. It is then explained how the same basic method can be used when estimating unsmooth probability densities and conditional variance functions.