4. PUBLICATIONS AND EDITING ACTIVITIES

The Institute publishes a Discussion Papers series and a Reprint series. The papers in both series are the output from the statistical research activities. Many collaborations (national and international) are going on with researchers from abroad. The following Discussion Papers and Reprints were issued during the period concerned by this report.

4.1 Discussion Papers

0301.

DE MACQ, I. and L. SIMAR, Hyper-rectangular space partitioning trees : a practical approach.

The process of computation of classification trees can be characterized as involving three basic choices : the type of splits considered in the growing process, the criterion to be optimized at each step of the process, and the way to get right-sized trees. Most implementations are ordinary binary trees, i.e. trees whose successive cuts are made by hyper-planes perpendicular to the axes, while most of the litterature concerns the various possible criteria and pruning methods. L. Devroye, L. Györfy and G. Lugosi (1996) define and consider the remarkable theoretical properties of a binary tree classifier whose prominent feature is the particular type of splits used in its construction : at a given node, partitioning is made by hyper-rectangles rather than hyper-planes. We propose an approximation of the solution for the complex optimization problem involved to allow insights on the practical advantages of those trees. Then we compare the performance of our algorithm with some leading algorithms for ordinary binary trees, namely CART and C4.5 as implemented in the Splus "tree" procedure and in SAS's Entreprise Miner respectively. For this purpose, data sets which traditionally enhance the weaknesses of classification trees are used, as well as data sets commonly used for comparisons.

0302.

CEBRIAN, A.C., DENUIT, M. and Ph. LAMBERT, Analysis of bivariate tail dependence using extreme value copulas : an application to the SOA medical large claims database.

The aim of this work is to analyze the dependence structure between losses and ALAE's relating to large claims using extreme value copulas. We propose a procedure to select and estimate the copula based on a parametric estimation of the dependence function. An application to the evaluation of reinsurance premiums is performed in group medical insurance. It clearly enhances the relevance of the approach.

0303.

PITREBOIS, S., DENUIT, M. and J-F. WALHIN, Setting a bonus-malus scale in the presence of other rating factors : Taylor's work revisited.

In this paper, we propose an analytic analogue to the simulation procedure described in Taylor (1997). We apply the formulas to a Belgian data set and discuss the interaction between a period and a posteriori ratemakings.

0304.

DENUIT, M. and S. LANG, Nonlife ratemaking with bayesian GAM's.

This paper aims to propose modern ratemaking techniques based on Generalized Additive Models (GAM's). The method accounts for discrete, continuous, categorical and spatial risk factors in a Bayesian framework. It uses computer-intensive simulation methods for statistical inference. Numerical illustrations based on a Belgian automobile portfolio enhance the interest of the approach.

0305.

DELOUILLE, V., JANSEN, M. and R. von SACHS, Second generation wavelet methods for denoising of irregularly spaced data in two dimensions.

We treat bivariate nonparametric regression, where the design of experiment can be arbitrarily irregular. Our method uses second-generation wavelets built with the lifting scheme: starting from a simple initial transform, we propose to use some predictor operators based on a generalization in two dimensions of the Lagrange interpolating polynomial. These predictors are meant to provide a smooth reconstruction. Next, we include an update step which helps to reduce the correlation amongst the detail coefficients, and hence stabilizes the final estimator. we use a Bayesian thresholding algorithm to denoise the empirical coefficients, and we show the performance of the resulting estimator through a simulation study.

0306.

BOUEZMARNI, T. and O. SCAILLET, Consistency of asymmetric kernel density estimators and smoothed histograms with application to income data.

We consider asymmetric kernel density estimators and smoothed histograms when the unknown probability density function f is defined on [0, + $\infty$). Uniform weak consistency on each compact set in [0, + $\infty$) is proved for these estimators when f is continuous on its support. Weak convergence in L₁ is also established. Finally we prove that the asymmetric kernel density estimator and the smoothed histogram converge in probability to infinity at x = 0 when the density is unbounded at x = 0. Monte Carlo results and an empirical study of the shape of a highly skewed income distribution based on a large micro­data set are finally provided.

0307.

SIMAR, L. and P.W. WILSON, Estimation and inference in two-stage, semi-parametric models of production processes.

plethora papers have used multi­stage estimation procedures where nonparametric estimates productive efficiency obtained the first stage and then regressed environmental variables subsequent stage in attempts account exogenous factors that might affect firms' performance. None these papers have described coherent data­generating process (DGP). Moreover, conventional approaches to inference employed these papers invalid due complicated, unknown serial correlation among estimated efficiencies. We first describe a DGP wherein firms' efficiencies influenced environmental variables. We then propose single a double bootstrap procedure; both permit valid inference, double bootstrap procedure improves statistical efficiency the second­stage regression. We examine the statistical performance of estimators using Monte Carlo experiments.

0308.

BOUEZMARNI, T. and J.M. ROLIN, Bernstein estimator for unbounded density function.

The nonparametric estimation for an unknown probability den­ sity function f with an known compact support [0, 1] not necessarily bounded at x = 0 is considered. For such class of density functions, we consider the Bernstein estimator. The uniform weak consistency and the uniform strong consistency on each compact I in (0, 1) are established for the Bernstein estimator. We prove also the almost sure convergence to infinity at x = 0 of the Bernstein estimator when the density function f is unbounded at x = 0. To select the optimal bandwidth parameter of the Bernstein estimator, the least squares cross­validation and the likelihood cross­validation methods are developed.

0309.

DENUIT, M. and J. DHAENE, Simple characterizations of comonotonicity and countermonotonicity by extremal correlations.

In this pedagogical note, it is shown how extremal values of classical measures of association like Pearson's correlation coefficient, Kendall's $\tau$ , Spearman's $\rho$ and Gini's $\gamma$, characterize comonotonicity and countermonotonicity. The link between zero­correlation and mutual independence is also examined.

0310.

VAN BELLEGEM, S. and R. von SACHS, Locally adaptive estimation of sparse evolutionary wavelet spectra.

We introduce a wavelet­based model of local stationarity. This model enlarges the class of locally stationary wavelet processes and contains processes whose spectral density function may change very suddenly in time. A notion of time­varying wavelet spectrum is uniquely defined as a wavelet­type transform of the autocovariance function with respect to so­called autocorrelation wavelets. This leads to a natural representation of the autocovariance which is localised on scales. One particularly interesting subcase arises when this representation is sparse, meaning that the nonstationary autocovariance process may be decomposed in the autocorrelation wavelet basis using few coefficients. We present a new test of sparsity for the wavelet spectrum. It is based on a non­asymptotic result on the deviations of a functional of a periodogram. The power of the test is discussed. We also present another application of this result given by the pointwise adaptive estimation of the wavelet spectrum. Properties of this estimator in homogeneous and inhomogeneous regions of the wavelet spectrum are studied.

0311.

PITREBOIS, S., DENUIT, M. and J.F. WALHIN, Fitting the Belgian Bonus-Malus system.

We show in this paper how to obtain the relativities of the Belgian Bonus­Malus System, including the special bonus rule sending the policyholders in the malus zone to initial level after four claim­free years. The model allows for a priori tarification. It is applied to a real­life portfolio.

0312.

GIJBELS, I., Inference for nonsmooth regression curves and surfaces using kernel-based methods.

In this paper we review kernel­based methods for detecting discontinuities in an otherwise smooth regression function or surface. In case of a possible discontinuous curve the interest might be in detecting the discontinuities, their jump sizes and finally to estimate the discontinuous curve. Alternatively, one might be uniquely interested in estimating directly the discontinuous curve preserving the jumps. A brief discussion on available kernel­based methods for testing for a continuous versus a discontinuous regression function, and for detecting discontinuities in regression surfaces is also provided.

0313.

DARAIO, C. and L. SIMAR, Introducing environmental variables in nonparametric frontier models : a probabilistic approach.

This paper proposes a general formulation of a nonparametric frontier model introducing external environmental factors that might influence the production process but are neither inputs nor outputs under the control of the producer. A representation is proposed in terms of a probabilistic model which defines the data generating process. Our approach extends the basic ideas from Cazals, Florens and Simar (2002) to the full multivariate case. We introduce the concepts of conditional efficiency measure and of conditional efficiency measure of order­m. Afterwards we suggest a practical way for computing the nonparametric estimators. Finally, a simple methodology to investigate the influence of these external factors on the production process is proposed. Numerical illustrations through some simulated examples and through a real data set on Mutual Funds show the usefulness of the approach.

0314.

VAN KEILEGOM, I., A note on the nonparametric estimation of the bivariate distribution under dependent censoring.

Consider the random vector (T₁, T₂), and assume that both T₁ and T₂ are subject to random right censoring. We propose new estimators of the bivariate and marginal distributions of T₁ and T₂. The estimators do not require the common assumption of independence between the vector of survival and censoring times, but allow for a certain type of dependent censoring. The proposed estimator of the marginal distribution generalizes the estimator of Cheng (1989). The estimators have intuitive, closed form expressions and are easy to compute. The weak convergence of the estimators is obtained. As an application we discuss the estimation of the regression coefficients in a polynomial regression model, when both the response and the covariate are subject to censoring.

0315.

PARK, B.U., SICKLES, R.C. and L. SIMAR, Semiparametric efficient estimation of dynamic panel data models.

This paper extends the semiparametric efficient treatment of panel data models pursued by Park and Simar (1994) and Park, Sickles, and Simar (1998, 2003) to a dynamic panel setting. We develop a semiparametric efficient estimator under minimal assumptions when the panel model contains a lagged dependent variable. We apply this new estimator to analyze the structure of demand between city pairs for selected U. S. airlines during the period 1979 I to 1992 IV.

0316.

BADIN, L. and L. SIMAR, Confidence intervals for DEA-type efficiency scores: how to avoid the computational burden of the bootstrap.

One important issue in statistical inference is to provide confidence intervals for the parameters of interest. Once the statistical properties of the estimators have been established, the corresponding asymptotic results can be used for constructing confidence intervals. However, in nonparametric efficiency estimation, the asymptotic properties of DEA estimators are only available for the bivariate case (Gijbels et al., 1999). An appealing alternative is the bootstrap method and a general methodology for applying bootstrap in nonparametric frontier estimation is provided by Simar and Wilson (1998, 2000b). Nevertheless, all the procedures involving bootstrap method are based on a large number of data replications, and in frontier estimation this approach also implies performing DEA (i.e. solving linear programs) a large number of times. Hence, a more simple and less computing intensive technique is always welcome. In this paper we propose a simple procedure for constructing confidence intervals for the efficiency scores. We consider some classical confidence intervals for an endpoint of a distribution and we show how these results can be adapted to the problem of frontier estimation. We provide an algorithm for constructing similar confidence intervals for the efficiency scores. Then some Monte Carlo experiments estimate the coverage probabilities of the obtained intervals. The results are quite satisfactory even for small samples. We then illustrate the approach with a real data set when analyzing the efficiency of 36 Air Controllers in Europe.

0317.

KNEIP, A., SIMAR, L. and P.W. WILSON, Asymptotics for DEA estimators in non-parametric frontier models.

Non­parametric data envelopment analysis (DEA) estimators based linear programming methods have been widely applied analyses productive efficiency. The distributions these estimators remain unknown except the simple case of input one output. This paper derives the asymptotic distribution DEA estimators under variable returns­to­scale. addition, bootstrap procedures (one based sub­sampling, the other based smoothing) shown provide consistent inference. smooth bootstrap requires smoothing irregularly­bounded density inputs and outputs well smoothing the DEA frontier estimate. Both bootstrap procedures allow dependence inefficiency process output levels and the mix inputs case input­oriented measures, inputs levels and mix of outputs the case output­oriented measures.

0318.

STEINMANN, L. and L. SIMAR, On the comparability of efficiency scores in nonparametric frontier models.

Data envelopment analysis (DEA) is widely used in the field of academic research, in business consulting and in a regulatory context. Usually it is the aim to estimate efficiency scores of decision making untis (DMU). The attempt to infer from a sample on the true, but unknown production technology males it a typical estimation procedure. Banker (1993) and Kneip, Park and Simar (1998) prove that the estimators obtained by DEA are biased, but under certain assumptions are consistent. Efficiency estimates obtained by DEA therefore seem to be suited for hypothesis testing, e.g. for comparison of mean efficiency between groups of observations. However, under certain circumstances - that will be analyzed in this paper - mean efficiency of groups of observations are biased to a different degree and thus differences in mean efficiency are also biased. Without bias correction, hypothesis tests of mean efficiencies between groups are then erroneous. In this paper an endicator is proposed to detect non-comparable mean efficiency scores. The procedure is illustrated in Monte Carlo simulations and applied to a real workd data set.

0319.

FRANCOIS, N., GOVAERTS, B. and B. BOULANGER, Optimal designs for inverse prediction in nonlinear calibration models.

Calibration models are intended to link a quantity of interest X (e.g. the concentration of a chemical compound) to a value Y obtained from a measurement device. In this context, a major concern is to build calibration models that are able to provide precise (inverse) predictions for X from measured responses Y. This paper aims at answering the following question : which experiments should be run to set up a (linear or nonlinear) calibration curve that maximises the inverse prediction precisions ? The well known class of optimal designs is presented as a possible solution. The calibration model setup is first reviewed in the linear case and extended to the heteroscedastic nonlinear one. In this general case, asymptotic variance and confidence interval formulae are derived for inverse predictions. Two optimality criteria are then introduced to quantify a priori the quality of inverse predictions for a given experimental design. The VI criterion is based on the integral of the inverse prediction variance over the calibration domain and the GI criterion on its maximum value. Algorithmic aspects of the optimal design generation are discussed. In a last section, the methodology is applied to 4 possible calibration models (linear, quadratic, exponential and four parameters logistic). VI and GI optimal designs are compared to classical D, V and G optimal designs. Their predictive quality is also compared to the one of simple traditional equidistant designs and it is shown that, even if these last designs have very different shapes, their predictive quality are not far from the optimal design ones. Finally, some simulations evaluate small sample properties of asymptotic inverse prediction confidence intervals.

0320.

DELAIGLE, A. and I. GIJBELS, Boundary estimation and estimation of discontinuity points in deconvolution problems.

This paper studies estimation of the boundary of the support of a density function when only a contaminated sample from the density is available. Estimation of the boundary of the support is a first necessary step when estimating a density with support different from the whole real line, since then modifications of the usual kernel type estimators are needed for consistent estimation of the density at the endpoints of its support. Apart from this, boundary estimation is also of in terest on its own, since it is the basic problem in, for example, frontier estimation in efficiency analysis in econometrics. The method proposed in this paper can also be used for estimating locations of discontinuity points of a density in the same deconvolution context. We establish the asymptotic law of the proposed estimator as well as approximate expressions for its mean squared error, and this for various types of error densities. These expressions then serve to discuss rates of convergence of the estimator and also shed some light on theoretical choices of the bandwidth parameter involved. An illustration of the method is given.

0321.

ALMEIDA, C. and M. MOUCHART, Identification of polychoric correlations : a copula approach.

The traditional model underlying the polychoric correlations among ordinal variables is revisited. This model relies on the idea of considering ordinal variables as discretization of corresponding continuous latent variables. The non­identification of the marginal distributions of the latent vector naturally leads to a copula approach; by so­doing, the role of the multivariate normality hypothesis of the latent vector is re­assessed.

0322.

ALMEIDA, C. and M. MOUCHART, A note on a copula approach to polychoric correlations.

Polychoric correlations among ordinal variables rest on the interpretation of the ordinal variables as discretization of latent continuous "ideally measured" variables and the assumption that these corresponding latent variables are jointly normally distributed. As the marginal distribution of these latent variables are arbitrary, for not being identified, the copula concept is a natural tool for the specification of the dependence structure. The role of the multivariate normality hypothesis is reexamined from a copulistic specification.

0323.

SIMAR, L., How to improve the performances of DEA/FDH estimators in the presence of noise?

In frontier analysis, most of the nonparametric approaches (DEA, FDH) are based on envelopment ideas which suppose that with probability one, all the observed units belong to the attainable set. In these "deterministic" frontier models, statistical theory is now mostly available. In the presence of noise, this is no more true and envelopment estimators could behave dramatically since they are very sensitive to extreme observations that could result only from noise. DEA/FDH techniques would provide estimators with an error of the order of the standard deviation of the noise. In this paper we propose to adapt some recent results on detecting change points, to improve the performances of the classical DEA/FDH estimators in the presence of noise. We show by simulated examples that the procedure works well when the noise is of moderate size, in term of noise to signal ratio. It turns out that the procedure is also robust to outliers.

0324.

SIMAR, L. and V. ZELENYUK, Statistical inference for aggregates of Farrell-type efficiencies.

In this study, we merge results directions in efficiency analysis research -the Aggregation and the Bootstrap- applied, as an example, to one of the most popular point-estimators of individual efficiency : the Data Envelopment Analysis (DEA) estimator. A natural context of the methodology developed here is a study of efficiency of a particular economic system (e.g., an industry) as a whole, or a comparison of efficiencies of distinct groups within such a system (e.g., private vs. public or regulated vs. non-regulated firms, etc.) Our methodology is justified by the (neo-classical) economic theory and is supported by carefully adapted statistical methods.

0325.

BIGOT, J., Landmark-based registration of 1D curves and functional analysis of variance with wavelets.

This paper is concerned with the problem of the alignment of multiple sets of curves and their comparison with FANOVA techniques. A nonparametric approach is proposed to estimate the zero­crossings lines of the continuous wavelet transform of a 1D signal observed with noise. A new tool, the "structural intensity", is introduced to represent the locations of the significant landmarks of an unknown curve via a probability density function. This technique yields an automatic landmark­based registration method to synchronize a set of curves. A fixed­effects FANOVA model is then used to test the significance of main/interaction effects and to show the usefulness of curve alignment. Some real examples arising from the biomedical area are used to illustrate the methodology.

0326.

DAOUIA, A. and L. SIMAR, Robust nonparametric estimators of monotone boundaries.

This paper revisits some asymptotic properties of the robust nonparametric estimators of order-m and order $\alpha$ quantile frontiers and proposes isotonized version of these estimators. Previous convergence properties of the order-m frontier are extended (from weak uniform convergence to complete uniform convergence). Complete uniform convergence of the order-m (and of the quantile order -$\alpha$) nonparametric estimators to the boundary is also established, for an appropriate choice of m (and of $\alpha$, respectively) as a function of the sample size. The new isotonized estimators share the asymptotic properties of the original ones and a simulated example shows, as expected, that these new versions are even more robust than the original estimators. The procedure is also illustated through a real data set.

0327.

VAN BELLEGEM, S. and R. von SACHS, On adaptive estimation for locally stationary wavelet processes and its applications.

The class of locally stationary wavelt is a wavelet-based model for covariance nonstationary zeromean time series. This paper presents an algorithm for the pointwise adaptive estimation of their time-varying spectral density. The performance of the procedure is evaluated on simulated and real times series. Two applications of the procedure are also presented and evaluated on real data. The first is a test of local significance for the coefficients of the so-called wavelet periodogram. The second is a new test of covariance stationarity.

0328.

OMBAO, H., von SACHS, R. and W. GUO, SLEX analysis of multivariate non-stationary time series.

We propose to analyze a multivariate non-stationary time series using the SLEX (Smooth Localized Complex EXponentials) library. The SLEX library is a collection of bases; each basis consists of the Slex waveforms which are orthogonal localized versions of the Fourier complex exponentials. In our procedure, we first build a family of multivariate SLEX models such that every model has a spectral representation in terms of a unique SLEX basis. The SLEX family provides a flexible representation for non-stationary random processes because every SLEX basis is localized in both time and frequency. The next step is to select a model using a penalized log energy criterion which we derive in this paper to be the Kullback-Leibler distance between a model and the empirical time series. In our procedure, we apply SLEX principal comonents analysis to obtain a decomposition of a possibly highly cross-correlated multivariate data set into non-stationary components with uncorrrelated (non-redundant) spectral information. The best model is then selected by computing the log energy criterion based on the SLEX principal components. The proposed SLEX analysis for multivariate non-stationary time series closely parallels traditional Fourier analysis of stationary time series. Hence, our method gives results that are easy to interpret. Moreover, the SLEX method uses computationally efficient algorithms and hence can easily handle massive data sets. We illustrate theSLEX method by its application to a multivariate brain waves data set recorded during an epileptic seizure.

0329.

MOUCHART, M. and M. VANDRESSE, A measure of market imperfection by frontier analysis.

In this paper, we propose an empirical method to measure the market imperfection and the bargaining power of the agents, by extending the methods of frontier analysis. A case study in the field of freight transport illustrates the proposed method.

0330.

MOUCHART, M. and J. ROMBOUTS, Clustered panel data models : an efficient approach for nowcasting from poor data.

Nowcasting regards the inference on the present realization of random variables, on the basis of information available until a recent past. This paper proposes a modelling strategy aimed at a best use of the data for nowcasting based on panel data with severe deficiencies, namely short times series and many missing data. The basic idea consists of introducing a clustering approach into the usual panel data model specification. A case study in the field of R&D variables illustrates the proposed modelling stategy.

0331.

BOUEZMARNI, T., MESFIOUI, M. and J-M. ROLIN, L₁ rate of convergence of asymmetric kernel density estimators and smoothed histograms.

The authors consider the Smoothed histograms (Gawronski and Stadtmuller (1980)) and the Gamma kernel density estimator (Chen 2000) for an iid sample of a density defined on [ 0, + $\infty$). They give the asymptotic behavior, the lower and upper bound of the expected average absolute error of this estimator on each compact. The general formula of the asymptotic behavior, the lower and upper bound of the expected average absolute error for the generalized kernel estimates will be established.

0332.

HEUCHENNE, C. and I. VAN KEILEGOM, Polynomial regression with censored data based on preliminary nonparametric estimation.

Consider the polynomial regression model Y = $\beta_{0}^{}$ + $\beta_{1}^{}$X + ... + $\beta_{p}^{}$X^p + $\sigma$(X)$\epsilon$, where $\sigma^{2}_{}$(X) = Var(Y| X) is unknown, and $\epsilon$ is independent of X and has zero mean. Suppose that Y is subject to random right censoring. A new estimation procedure for the parameters $\beta_{0}^{}$,...,$\beta_{p}^{}$ is proposed, which extends the classical least squares procedure to censored data. The proposed method is inspired by the method of Buckley and James (1979), but is, unlike the latter method, a non­iterative procedure due to nonparametric preliminary estimation of the conditional regression function. The asymptotic normality of the estimators is established. Simulations are carried out for both methods and they show that the proposed estimators have usually smaller variance and smaller mean squared error than the Buckley­James estimators. The two estimation procedures are also applied to a medical and an astronomical data set.

0333.

EINMAHL, J.H.J. and I. VAN KEILEGOM, Goodness-of-fit tests in nonparametric regression.

Consider the nonparametric regression model Y = m(X) + $\epsilon$, where the function m is smooth, but unknown, and $\epsilon$ is independent of X. We construct omnibus goodness­of­fit tests, based on n independent copies of (X, Y), for the independence of$\epsilon$ and X and establish asymptotic results for the proposed tests statistics. We investigate their finite sample properties through a simulation study and present an econometric application to household data. One testing procedure is based on differences of neighboring Y 's, whereas the other one makes use of an estimator of m. The proofs are based on delicate weighted empirical process theory.

0334.

GIJBELS, I., Monotone regression.

In nonparametric regression the objective is to explore the relationship between an explanatory variable and the variable of interest. Sometimes it is plausible to assume that the regression function is monotone. We review nonparametric methods for estimating a monotone regression function, as well as testing procedures for testing for monotonicity. A brief discussion on related estimation and testing problems is given.

4.2 Consulting Reports

CR0301

MOUCHART, M. and J. ROMBOUTS, Econometric models for nowcasts on R & D variables, subcontracting from CAMIRE (Luxembourg) for Eurostat.

CR0302

MOUCHART, M. and L. SIMAR, Efficiency analysis of air navigation services provision (II): further insights, consulting report for Eurocontrol, Brussels.

CR0303

MOUCHART, M. and J. ROMBOUTS, Evaluating and updating econometric Models for Nowcasts on R&D variables, subcontracting from CAMIRE (Luxembourg) for Eurostat.

CR0304

VAN BELLEGEM, S. and P. VANDEN EECKAUT, Assessment of the concentration level of chemical substances in river network - Part V - A model for trend estimation adapted to monitoring data, study for EuroChlor.

CR0305

DELAIGLE, A., GOVAERTS, B. et J. HOEFFELMAN (Elia), Etude des champs magnétiques sous les lignes à haute tension de Belgique.

4.3 Published Papers

184.

MOUCHART, M. and E. SAN MARTIN. Specification and identification issues in models involving a latent hierarchical structure. Journal of Statistical Planning and Inference, 111, 143-163, 2003.

186.

DONOHO, D.L., MALLAT, S., von SACHS, R. and Y. SAMUELIDES. Locally stationary covariance and signal estimation with macrotiles. IEEE Transactions on Signal Processing, 51, 3, 614-627, 2003.

188.

DENUIT, M., LEFEVRE, Cl. and Ph. PICARD. Polynomial structures in order statistics distributions. Journal of Statistical Planning and Inference, 113, 151-178, 2003.

189.

VANDENHENDE, F. and Ph. LAMBERT. Improved rank-based dependence measures for categorical data. Statistics & Probability Letters, 63, 157-163, 2003.

190.

PURCARU, O. and M. DENUIT. Dependence in dynamic claim frequency credibility models. Astin Bulletin, 33, 1, 23-40, 2003.

191.

BROUHNS, N. and M. DENUIT. Actuarial modelling of longitudinal claims data through GAMM's : some methodological results. Deutsche Gesellschaft für Versicherungs-und Finanzmathematik e.V, 26, 1, 25-39, 2003.

192.

CEBRIAN, A.C., DENUIT, M. and Ph. LAMBERT. Generalized pareto fit to the society of actuaries' large claims database. North American Actuarial Journal, 7, 3, 18-36, 2003.

193.

FROGNIER, A-P. et M. MOUCHART. La wallonie : l'impact des positions sociales, des clivages et des enjeux sur le vote en 1999. Dans Elections : La rupture? Le comportement des Belges face aux élections de 1999. A-P. Frognier et A-M. Aish, Bruxelles : De Boeck, 13-27, 2003.

194.

VANDENHENDE, F., LAMBERT, Ph. and N. RAMADAN. Statistical models for the analysis of controlled trials on acute migraine. Pharmaceutical Statistics, 2, 199-210, 2003.

195.

FLAHAUT, B., MOUCHART, M., SAN MARTIN, E. and I. THOMAS. The local spatial autocorrelation and the kernel method for identifying black zones. A comparative approach. Accident Analysis and Prevention, 35, 991-1004, 2003.

196.

OULHAJ, A. and M. MOUCHART. Partial sufficiency with connection to the identification problem. METRON. International Journal of Statistics, LXI, 2, 267-283, 2003.

197.

SIMAR, L. Detecting outliers in frontier models : a simple approach. Journal of Productivity Analysis, 20, 391-424, 2003.

198.

GIJBELS, I. Inference for nonsmooth regression curves and surfaces using kernel-based methods. Recent Advances and Trends in Nonparametric Statistics, 183-201, 2003.

199.

ZHANG, J. and I. GIJBELS. Sieve empirical likelihood and extensions of the generalized least squares, Scandinavian Journal of Statistics, 30, 1-24, 2003.

200.

HALL, P. and I. VAN KEILEGOM. Using difference-based methods for inference in nonparametric regression with time series errors, Journal of Royal Statistical Society, Series B, 65, Part 2, 443-456, 2003.

201.

AKRITAS, M.G. and I. VAN KEILEGOM. Estimation of bivariate and marginal distributions with censored data, Journal of Royal Statistical Society, Series B, 65, Part 2, 457-471, 2003.

202.

DU, Y., AKRITAS, M.G. and I. VAN KEILEGOM. Nonparametric analysis of covariance for censored data, Biometrika, 90, 2, 269-287, 2003.

203.

CHEN, X., LINTON, O. and I. VAN KEILEGOM. Estimation of semiparametric models when the criterion function is not smooth, Econometrica, 71, 5, 1591-1608, 2003.

204.

PITREBOIS, S., DENUIT, M. and J.-F. WALHIN. Setting a bonus-malus scale in the presence of other rating factors: Taylor's work revisited. Astin Bulletin. 33, 2, 419-436, 2003.

205.

PITREBOIS, S., DENUIT, M. and J.-F. WALHIN. Tarification automobile sur données de panel. Mitteilungen der Schweiz Aktuarvereinigung. 1, 51-81, 2003.

206.

PITREBOIS, S., DENUIT, M. and J.-F. WALHIN. Fitting the Belgian Bonus-Malus system. Belgian Actuarial Bulletin. 3, 1, 58-62, 2003.

207.

CEBRIAN, A., DENUIT, M. and Ph. LAMBERT. Analysis of bivariate tail dependence using extreme value copulas: an application to the SOA medical large claims database. Belgian Actuarial Bulletin. 3, 1, 33-41, 2003.

208.

DENUIT, M. and J. DHAENE. Simple characterizations of comonotonicity and countermonotonicity by extremal correlations. Belgian Actuarial Bulletin. 3, 1, 22-27, 2003.

209.

DELWARDE, A. and M. DENUIT. Importance de la période d'observation et des âges considérés dans la projection de la mortalité selon la méthode de Lee-Carter. Belgian Actuarial Bulletin. 3, 1, 1-21, 2003.

210.

BROUHNS, N., GUILLEN, M., DENUIT, M. and J. PINQUET. Bonus-malus scales in segmented tariffs with stochastic migration between segments. The Journal of Risk and Insurance. 70, 4, 577-599, 2003.

211.

DENUIT, M., LEFEVRE, C. and M. MESFIOUI. On spline approximation for bivariate functions of increasing convex type. Revue d'Analyse Numérique et de Théorie de l'Approximation. 32, 2, 145-159, 2003.

212.

GIJBELS, I. and U. GURLER. Estimation of a change point in a hazard function based on censored data . Lifetime Data Analysis. 9, 395-411, 2003.

213.

FRYZLEWICZ, P., VAN BELLEGEM, S. and R. von SACHS. Forecasting non-stationary time series by wavelet process modelling. The Annals of the Institute of Statistical Mathematics. 55, 4, 737-764, 2003.

215.

CLAESKENS, G. and I. VAN KEILEGOM. Bootstrap confidence bands for regression curves and their derivatives, The Annals of Statistics, 31, 6, 1852-1884, 2003.

216.

PARK, B., SICKLES, R. and L. SIMAR. Semiparametric efficient estimation of AR(1) panel data models, Journal of Econometrics, 117, 2, 279-311.

217.

VAN BELLEGEM, S., FRYZLEWICZ, P. and R. von SACHS. A wavelet-based model for forecasting non-stationary processes. Inst. Phys. Conf. Ser. 173, 955-958, 2003. Paper presented at 24th Int. Coll. Group Theoretical Methods in Physics, Paris, France, July 2002.

218.

TILQUIN, P., VAN KEILEGOM, I., COPPIETERS, W., LE BOULENGE, E. and P.V. BARET. Non-parametric interval mapping in half-sib designs : use of midranks to account for ties. Genet. Res., Camb. 81, 221-228, 2003.

219.

CLAESKENS, G. and I. VAN KEILEGOM, Bootstrap confidence bands for regression curves and their derivatives. The Annals of Statistics. 31, 6, 1852-1884, 2003.

4.4 Books published by members of the Institute

FLORENS, J.P., MOUCHART, M. and J.M. ROLIN. Elements of Bayesian Statistics, 544 pp, Marcel Dekker: New York, 1990.

HÄRDLE, W. and L. SIMAR (editors). Computer Intensive Methods in Statistics, 175 pp, Statistics and Computing, I, Physica-Verlag: Berlin, 1993.

HÄRDLE, W., KLINKE, S. and B.A. TURLACH. XploRe: An Interactive Statistical Computing Environment, 387 pp, Statistics and Computing, Springer-Verlag: New York, 1995.

FAN, J. and I. GIJBELS. Local Polynomial Modelling and its Applications, 341 pp, Chapman and Hall: London, 1996.

KAAS, R., GOOVAERTS, M.J., DHAENE, J., and M. DENUIT. Modern Actuarial Risk Theory, Kluwer Academic Publishers: Dordrecht, 2001.

WUNSCH, G., MOUCHART, M. and J. DUCHÊNE (editors). The Life Table : Modelling Survival and Death, book series : European Studies of Population, vol. 11, Kluwer Academic Publishers : Dordrecht, 2002.

HÄRDLE, W. and L. SIMAR. Applied Multivariate Statistical Analysis, 486 pp., Springer Verlag: Berlin, 2003.

4.5 Editing activities

Michel DENUIT

Proceedings Editor for Insurance: Mathematics and Economics
Editor of Belgian Actuarial Bulletin
Associate Editor Australian and New-Zeeland Journal of Statistics
Member of the Advisory Board of the Wiley Encyclopedia of Actuarial Science.

Irène GIJBELS

(Associate) Editor of Journal of Multivariate Analysis.
Associate Editor of Journal of Computational and Graphical Statistics.
Associate Editor of Statistica Sinica.

Philippe LAMBERT

Co-editor of B-Stat News.

Léopold SIMAR

Associate Editor of Journal of Productivity Analysis

Ingrid VAN KEILEGOM

Associate Editor of Journal of the Royal Statistical Society - Series B