In contrast to previous studies on firm survival which tend to focus on features related to the structure of the firms and their area of activity, our aim here is to widen the perspective usually adopted in the field, taking into account a larger and more qualitative set of variables. Among these variables, features related to the individual characteristics of the entrepreneur, to the context of entrepreneurship and to the insertion in entrepreneurial networks are significant to explain the life span of new firms. The empirical material is drawn from two surveys, which provide detailed data about a group of new firms created in France in 1994 and closed down before 1997 or still running in 1997. Our empirical approach on qualitative data is based on data analysis methods (linear discriminant analysis, barycentric discriminant analysis, analysis of variance). According to the characteristics of the entrepreneur, the main explanatory factors for the survival of new firms are the fact that they are entrepreneurs who have taken over firms, that they have acquired during their previous occupational activity an experience in the same branch of activity and that they experience a successful integration into the entrepreneurial networks. These three factors show that the survival of young firms is indirectly conditioned by the existence of an initial custom, by the mastery of a job and by the know-how in the entrepreneurial function.

Although surface electromyography (sEMG) has a high correlation to muscle force, an accurate model that can estimate joint torque from sEMG is still elusive. Artificial neural networks (NN), renowned as universal approximators, have been employed to capture this complex nonlinear relation. This work focuses on investigating possible improvements to the NN methodology and algorithm that would consistently produce reliable sEMG-to-knee-joint torque mapping for any individual. This includes improvements in number of inputs, data normalization techniques, NN architecture and training algorithms. Data (sEMG) from five knee extensor and flexor muscle from one subject were recorded on 10 random days over a period of 3 weeks whilst subject performed both isometric and isokinetic movements. The results indicate that incorporating more muscles into the NN and normalizing the data at each isometric angle prior to NN training improves torque estimation. The mean lowest estimation error achieved for isometric motion was 10.461% (1.792), whereas the lowest estimation errors for isokinetic motion were larger than 20%.

No abstract

No abstract

No abstract

No abstract

L’objet de cet article est de rechercher les facteurs qui différencient les chargés de clientèle des banques mutualistes des chargés de clientèle des banques commerciales en matière de financement des PME. Ces facteurs de différenciation sont identifiés à partir des réponses obtenues dans le cadre d’une enquête auprès d’un échantillon représentatif selon le type de banque de la population des chargés de clientèle PME de la région Basse-Normandie. Les méthodes d’analyse des données réalisées mettent en évidence l’importance de l’ancienneté de la relation pour les banquiers mutualistes à l’opposé des banquiers SA qui prennent davantage en compte l’étude des documents comptables et l’analyse de la faisabilité économique du projet de l’entreprise. Classification JEL : G21, D82, L31, C42

We consider an industry including both leveraged and unleveraged firms engaged in a sequential decision-making process. At the first stage a firm and its bank, considering the production cost uncertainty and the probability distribution of the random shock, evaluate a bankruptcy risk; at the second stage firms engage in Cournot competition. By introducing an additional upstream stage we examine how incentives to merge with competitors are altered when shareholders of leveraged firms are protected by the limited liability. We demonstrate that a merger increases the probability of bankruptcy for the merged firm if the merger involves only leveraged firms, but this probability decreases if the merger involves at least one unleveraged firm. The welfare loss associated with anticompetitive effects of mergers is lower when the coalition gathers unleveraged firms rather than leveraged ones. Consequently, we argue that in evaluating proposed mergers Competition Authorities should take into account the financial structure of both merging firms and outsiders.