Two-stage Huber estimation

In this paper we propose a new robust estimator in the context of two-stage estimation methods directed towards the correction of endogeneity problems in linear models. Our estimator is a combination of Huber estimators for each of the two stages, with scale corrections implemented using preliminary median absolute deviation estimators. In this way we obtain a two-stage estimation procedure that is an interesting compromise between concerns of simplicity of calculation, robustness and efficiency. This method compares well with other possible estimators such as two-stage least-squares (2SLS) and two-stage least-absolute-deviations (2SLAD), asymptotically and in finite samples. It is notably interesting to deal with contamination affecting more heavily the distribution tails than a few outliers and not losing as much efficiency as other popular estimators in that case, e.g. under normality. An additional originality resides in the fact that we deal with random regressors and asymmetric errors, which is not often the case in the literature on robust estimators.     

The symmetric and asymmetric Choquet integrals on finite spaces for decision making

In this paper, we give a mathematical analysis of symmetric and asymmetric Choquet integrals in the view of decision making in a finite setting. These integrals present two ways of dealing with negative integrands. The analysis is done with the aid of the Möbius and interaction transforms, this last one having an interesting interpretation in multicriteria decision making (MCDM). The last part of the paper shows the application of these two integrals in MCDM.     

Optimal pension funding with demographic instability and endogenous returns on investment

This paper tries to explore some optimal funding policies for pension systems in a general equilibrium setting where funding affects returns on investment and wages through its impact on capital formation. This is done in the context of irregular demographic evolutions such as those expected in developed countries for the next century. Particular attention is given to the intergenerational welfare criterion which is used for designing optimal policies. It appears that funding receives low justification with a welfare criterion which assumes a high substitutability between consumptions of successive cohorts, implying a low concern for intergenerational equity. Funding is highly justified in the opposite case where a high level of consumption for some cohorts is not considered as a compensation for low consumption by others. However the optimal patterns of transfers and savings which are found in this latter case are not straightforward. Some simpler funding rules are explored in the last section of this paper, which show that non-optimal funding may imply, on the contrary, a high level of inequality between subsequent generations.Paper presented at the ISPE-conference Fiscal implications of an ageing population, May–June 1990 at Vaalsbroek, The Netherlands. We thank Pierre Pestieau for his remarks on a first draft of this paper. We also thank for their comments our two discussants Carol Propper and Lans Bovenberg, as well as other participants to the ISPE Conference. Any errors are, of course, ours.     

Stable and visualizable Gaussian parsimonious clustering models

Gaussian mixture model-based clustering is now a standard tool to determine a hypothetical underlying structure in continuous data. However, many usual parsimonious models, despite either their appealing geometrical interpretation or their ability to deal with high dimensional data, suffer from major drawbacks due to scale dependence or unsustainability of the constraints after projection. In this work we present a new family of parsimonious Gaussian models based on a variance-correlation decomposition of the covariance matrices. These new models are stable when projected into the canonical planes and, so, faithfully representable in low dimension. They are also stable by modification of the measurement units of the data and such a modification does not change the model selection based on likelihood criteria. We highlight all these stability properties by a specific graphical representation of each model. A detailed Generalized EM (GEM) algorithm is also provided for every model inference. Then, on biological and geological data, we compare our stable models to standard ones (geometrical models and factor analyzer models), which underlines all the profit to obtain unit-free models.     

Introduction. Ce que les écrits font au travail

As is known, large numbers of writings circulate within every kind of professional environment. As a result, there is extensive research on the formats, uses and roles of writings in work. However, the approach taken here is not to describe what is done with writings in work. The central hypothesis rests on a performative vision of writings, one that gives them the power to act, agency. It can be summed up as follows: what do writings “do” to work? Studying work through its writings is a way to connect two dimensions that are too often kept separate: situated activities on the one hand, organizational functions on the other. Indeed, writings are material and robust — though not always lasting — traces of concrete activities, and they provide a handhold for the rigorous analysis of work as close as possible to its conditions of realization. However, writings travel and circulate. They form links between workers, groups, workshops, services, firms, without which production could not take place. They stretch work beyond its source. To track them is therefore to shift the field of enquiry to moments other than the present of the observed situation, and also to other nearby, surrounding workspaces. This brings out the procedural and interactive dimensions of organizations, their integrating capacity. Exploring what writings do to work is therefore not only to place oneself at an intermediate level between situated action and organized action. For the sociologist, it is a way of connecting these two perspectives “on paper”, by embedding the analysis of the actual work in spatially and temporally broader collective actions, and by documenting organizational analysis on the basis of observed work situations. It is a way of contributing to a more extensive ecology of activity, more clearly anchored within its organizational envelope.     

A general central limit theorem and a subsampling variance estimator for α‐mixing point processes

We establish a central limit theorem for multivariate summary statistics of nonstationary α‐mixing spatial point processes and a subsampling estimator of the covariance matrix of such statistics. The central limit theorem is crucial for establishing asymptotic properties of estimators in statistics for spatial point processes. The covariance matrix subsampling estimator is flexible and model free. It is needed, for example, to construct confidence intervals and ellipsoids based on asymptotic normality of estimators. We also provide a simulation study investigating an application of our results to estimating functions.     

Factors that Determine or Influence Managerial Innovation in Public Contexts: The Case of Local Performance Management

The use of Performance-Based Management (PBM) by French public organizations has sharply accelerated over the last few years. Moreover, very few studies have analyzed this change. The object of this article is thus to conduct an in-depth analysis of the dynamics of managerial innovation in three French local authorities via the study of the factors that foster the adoption and implementation of PBM by this type of organization. We demonstrate a tendency to overestimate the explanatory power of the neo-institutionalist theories. Indeed, our analysis illustrates a balanced vision of the human, technical and contextual factors at the origin of this type of managerial innovation.