In a first time we draw a rough shape of a general formal framework for polynomial approximation theory which encompasses the existing one by allowing the expression of new types of results. We show how this framework incorporates all the existing approximation results and, moreover, how new...

Lexical parsing is the whole set of techniques to increase from a raw text to a sequence of tagged words ; these tags are morphological and grammatical information. Lexical parsing generally uses an electronic dictionary. But a few of them accommodate themselves to proper nouns that are a...

Une methodologie alternative aux modeles multicriteres fondes sur l'etude economique et financiere de la firme est proposee afin d'estimer le risque de defaut des entreprises et d'elaborer un indicateur du risque de faillite. Nous recourons a une methode de Gestion Actif - Passif d'estimation...

En traitement automatique de la langue naturelle, l'analyse lexicale d'un texte est faite a partir de dictionnaires electroniques. Ceux-ci comportent tres peu de Noms Propres. Un travail de collecte des toponymes (noms de lieux) et gentiles (noms d'habitants) pour la France a ete mene. 55508...

In this paper, we study almost periodic (a.p.) solutions of discrete dynamical systems. We first adapt some results on a.p. differential equations to a.p. difference equations, on the link between boundedness of solutions and existence of a.p. solutions.

Cumulative prospect theory was introduced by Tversky and Kahneman to combine the empirical realism of their original prospect theory with the theoretical advantages of rank-dependent utility. Preference axiomatizations were provided in several papers. All exosting axiomatizations, however, only...

This working paper details unit root tests for monthly series, with the possibility of the stationary part having been generated by an ARMA process. The paper is complemented by an application to the four basic stochastic components of the consumer price index. Test evidences that the long-run...

Using interpolation of the Spanish domestic private consumption deflactor, we review some of the signals extraction techniques, as well as interpolation of time series, i.e. deriving quarterly data from annual data. For the decomposition of a series in its observable components (trend plus...

We present a new Fair Queuing policy adapted to low-bandwidth flows, and preventing discrepancy which is encountered under almost all of the existing Fair Queuing policies. Unlike Packet-by-packet Generalized Processos Sharing PGPS and its version Worst-case Fair, Fair Queuing WF2Q which...

The paper deals with the problem of identifying stochastic unobserved two-component models, as in seasonal adjustment or trend-cycle decompositions. Solutions based on the properties of the unobserved component estimation error are considered, and analytical expressions for the variances and...