IMATH

Résumé: Dans ce travail, nous montrons comment les algorithmes PLS, correctement ajustés, peuvent travailler comme des algorithmes de Codage Optimal. Cette nouvelle fonctionnalité du PLS, qui avait été jusqu’à maintenant totalement inexplorée, nous a permis de mettre au point une nouvelle série de méthodes PLS: les méthodes Non-Metric PLS (NM-PLS). Abstract: In this work we find out how PLS algorithms, properly adjusted, can work as optimal scaling algorithms. This new feature of PLS, which had until now been totally unexplored, allowed us to devise a new suite of PLS methods: the Non-Metric PLS (NM-PLS) methods. Mots-clès: Analyse des données- data mining, Problèmes inverses et sparsité

This book series features volumes composed of select contributions from workshops and conferences in all areas of current research in mathematics and statistics, including OR and optimization. In addition to an overall evaluation of the interest, scientific quality, and timeliness of each proposal at the hands of the publisher, individual contributions are all refereed to the high quality standards of leading journals in the field. Thus, this series provides the research community with well-edited, authoritative reports on developments in the most exciting areas of mathematical and statistical research today.

L'approche PLS aux modeles a equations structurelles (PLS Path Modeling, PLS-PM) est couramment consideree comme une approche basee sur les composantes. Cette methode a ete recemment revisitee en tant que cadre general pour l'analyse des tableaux multiples. Nous proposons ici deux nouvelles methodes d'estimation des poids externes dans le cadre de la PLS-PM: le Mode PLScore et le Mode PLScow. Chaque mode est fonde sur l'utilisation de la regression PLS pour l'etape d'estimation externe. Toutefois, en Mode PLScore une regression PLS est executee sous les contraintes classiques de la PLS-PM de variance unitaire pour les scores des variables latentes ; tandis que dans le Mode PLScow les poids externes sont contraints d'avoir une norme unitaire. Cette derniere contrainte est la contrainte classique de normalisation dans le cadre de la regression PLS. Nous montrons comment les deux nouveaux modes sont lies aux methodes d'estimation externe classiques de la...

This paper presents an original methodology to estimate delay risk a few days before operations with generalized linear models. These models represent a given variable with any distribution from the exponential family, allowing to compute for any subject its own probability distribution according to its features. This methodology is applied on small delays (less than 20 minutes) of high-speed trains arriving at a major french station. Several distributions are tested to fit delay data and three scenarios are evaluated: a single GLM with a negative binomial distribution and two two-part models using both a logistic regression as first part to compute the probability of arriving on time, and a second part using a negative binomial or a lognormal distribution to obtain the probabilities associated with positive delay values. This paper also proposes a validation methodology to assess the quality of these probabilistic prediction based on two aspects: calibration and discrimination.

After a presentation of the classical methodology of conjoint analysis we advocate the use of PLS regression instead of classical regression in order to obtain a better accuracy in the estimation of utilities. PLS regression provides also interesting graphical representations allowing quick and easy interpretation of the relations among preferences, attributes and profiles.

Nowadays there is a pre-eminent need to measure very complex phenomena like poverty, progress, well-being, etc. As is well known, the main feature of a composite indicator is that it summarizes complex and multidimensional issues. Thanks to its features, Structural Equation Modeling seems to be a useful tool for building systems of composite indicators. Among the several methods that have been developed to estimate Structural Equation Models we focus on the PLS Path Modeling approach (PLS-PM), because of the key role that estimation of the latent variables (i.e. the composite indicators) plays in the estimation process. In this paper we provide a suite of statistical methodologies for handling categorical indicators with respect to the role they have in a system of composite indicators. A categorical variable can play an active or a moderating role. An active categorical variable directly participates in the construction of the model. In other words, it is a categorical indicator im...

This paper deals with robustness evaluation at station, and in particular for the train plat-forming problem (TPP). This problem consists in a platform and route assignment in station for each scheduled train. A classical robustness evaluation is simulation: simulated delays are injected on arriving and departing trains then propagated, and results are averaged on a large number of trials. A robust solution of the TPP aims to limit the total amount of secondary delays. However, a simulation framework at station is difficult t o c alibrate: it requires a realistic delays generator and an accurate operating rules modeling. This paper proposes an original simulation framework using classical statistical learning algorithms and calibration assessment methods to model simulation inputs. This methodology is applied on delay data to simulate delay propagation at station. It highlights the importance of delay calibration by showing that even slight miscalibration of inputs can lead to stron...

Mots-clés: Analyse des données- data mining; Problèmes inverses et sparsité Résumé: L’approche PLS aux modèles à équations structurelles (PLS Path Modeling, PLS-PM) est couramment considérée comme une approche basée sur les composantes. Cette méthode a été récemment revisitée en tant que cadre général pour l’analyse des tableaux multiples. Nous proposons ici deux nouvelles méthodes d’estimation des poids externes dans le cadre de la PLS-PM: le Mode PLScore et le Mode PLScow. Chaque mode est fondé sur l’utilisation de la régression PLS pour l’étape d’estimation externe. Toutefois, en Mode PLScore une régression PLS est exécutée sous les contraintes classiques de la PLS-PM de variance unitaire pour les scores des variables latentes; tandis que dans le Mode PLScow les poids externes sont contraints d’avoir une norme unitaire. Cette dernière contrainte est la contrainte classique de normalisation dans le cadre de la régression PLS. Nous montrons comment les deux nouveaux modes sont liés...

We propose a multidimensional instrument to assess the degree of presence of marketing capabilities a firm possesses, at three levels of abstraction. We first present the theoretical framework for marketing capabilities and discuss the main scales proposed by Vorhies et al. Then, we detail the steps required to develop and validate our third-order formative instrument. We assess the convergent and discriminant validity of the proposed instrument via partial least squares path modelling (PLS-PM) applied to a sample of 199 French small- and medium-sized enterprises (SMEs). Finally, we check the nomological validity of our instrument by testing the positive effect of marketing capabilities on organisational performance.

Notre ambition est de proposer un instrument multidimensionnel permettant de décrire le degré de présence des principales capacités marketing sur trois niveaux d’abstraction. Après avoir présenté le cadre théorique relatif aux capacités marketing, l’article souligne tout d’abord les limites des principales échelles proposées par Vorhies et al. (1999 ; 2009), Vorhies et Harker (2000), et Vorhies et Morgan (2003 ; 2005). Ensuite, les étapes nécessaires au développement et à la validation d’un index multidimensionnel formatif de troisième ordre sont détaillées. Sur la base d’une collecte de données réalisée auprès d’un échantillon de 199 PME françaises, la phase d’analyse de la validité convergente et discriminante de l’instrument est réalisée à l’aide de l’approche PLS aux modèles à variables latentes (PLS-PM). Enfin, la validité nomologique de l’instrument proposé est confirmée via l’étude de l’influence des capacités marketing sur la performance organisationnelle.

Partial least squares (PLS) refers to a set of iterative algorithms based on least squares that implement a broad spectrum of both explanatory and exploratory multivariate techniques, from regression to path modeling, and from principal component to multi-block data analysis. This article focuses on PLS regression and PLS path modeling, which are PLS approaches to regularized regression and to predictive path modeling. The computational flows and the optimization criteria of these methods are reviewed in detail, as well as the tools for the assessment and interpretation of PLS models. The most recent developments and some of the most promising on going researches are enhanced.