Influence analysis in multivariate linear general models

In this paper we obtain several influence measures for the multivariate linear general model through the approach proposed by Muñoz-Pichardo et al. (1995), which is based on the concept of conditional bias. An interesting charasteristic of this approach is that it does not require any distributional hypothesis. Appling the obtained results to the multivariate regression model, we obtain some measures proposed by other authors. Nevertheless, on the results obtained in this paper, we emphasize two aspects. First, they provide a theoretical foundation for measures proposed by other authors for the mul¬tivariate regression model. Second, they can be applied to any linear model that can be formulated as a particular case of the multivariate linear general model. In particular, we carry out an application to the multivariate analysis of covariance.     

The Ranking of the Regions With Regard to Their Sports Facilities to Improve Their Planning in Sport: The Case of Spain

Sports facility planning by the public authorities should be considered to be one of the main policies to have an indirect influence on the aim of ensuring a healthy population, as well as reducing its physical inactivity. Our research aims to study and compare the state of sports infrastructures in the regions of Spain, using a synthetic indicators. Our results should serve not only to detect the vast imparity between different regions, which has come about because of varying investment and management with regard to sports infrastructures, but also to redirect building planning and sports management policies onto the right path, and also give food for thought.     

Approximating probability density functions in hybrid Bayesian networks with mixtures of truncated exponentials

Mixtures of truncated exponentials (MTE) potentials are an alternative to discretization and Monte Carlo methods for solving hybrid Bayesian networks. Any probability density function (PDF) can be approximated by an MTE potential, which can always be marginalized in closed form. This allows propagation to be done exactly using the Shenoy-Shafer architecture for computing marginals, with no restrictions on the construction of a join tree. This paper presents MTE potentials that approximate standard PDF’s and applications of these potentials for solving inference problems in hybrid Bayesian networks. These approximations will extend the types of inference problems that can be modelled with Bayesian networks, as demonstrated using three examples.     

Bivariate income distributions with lognormal conditionals

In this paper, the most general bivariate distribution with lognormal conditionals is fully characterized, using the methodology proposed by [3]. The properties of the new family are studied in detail, including marginal and conditional distributions, regression functions, dependence measures, moments and inequality measures. The new distribution is very broad, and contains as a particular case the classical bivariate lognormal distribution. Several subfamilies are studied and a generalization of the basic model is discussed. Finally, we present an empirical application. We estimate and compare the basic model proposed in the paper with a classical model, using data from the European Community Household Panel in different periods of time.     

Testing the Martingale Difference Hypothesis

In this paper we consider testing that an economic time series follows a martingale difference process. The martingale difference hypothesis has typically been tested using information contained in the second moments of a process, that is, using test statistics based on the sample autocovariances or periodograms. Tests based on these statistics are inconsistent since they cannot detect nonlinear alternatives. In this paper we consider tests that detect linear and nonlinear alternatives. Given that the asymptotic distributions of the considered tests statistics depend on the data generating process, we propose to implement the tests using a modified wild bootstrap procedure. The paper theoretically justifies the proposed tests and examines their finite sample behavior by means of Monte Carlo experiments.     

Analysis of individual- and time-specific covariate effects on survival of Serinus serinus in north-eastern Spain

We developed models for the analysis of recapture data for 2678 serins ( Serinus serinus ) ringed in north-eastern Spain since 1985. We investigated several time- and individual-specific factors as potential predictors of overall mortality and dispersal patterns, and of gender and age differences in these patterns. Time-specific covariates included minimum daily temperature, days below freezing, and abundance of a strong competitor, siskins ( Carduelis spinus ) during winter, and maximum temperature and rainfall during summer. Individual covariates included body mass (i.e. body condition), and wing length (i.e. flying ability), and interactions between body mass and environmental factors. We found little support of a predictive relationship between environmental factors and survival, but good evidence of relationships between body mass and survival, especially for juveniles. Juvenile survival appears to vary in a curvilinear manner with increasing mass, suggesting that there may exist an optimal mass beyond which increases are detrimental. The mass-survival relationship does seem to be influenced by at least one environmental factor, namely the abundance of wintering siskins. When siskins are abundant, increases in body mass appear to relate strongly to increasing survival. When siskin numbers are average or low the relationship is largely reversed, suggesting that the presence of strong competition mitigates the otherwise largely negative aspects of greater body mass. Wing length in juveniles also appears to be related positively to survival, perhaps largely due to the influence of a few unusually large juveniles with adult-like survival. Further work is needed to test these relationships, ideally under experimentation.     

Evaluation of vectorization/parallelization techniques: application to nonparametric curve estimation

The simulation of statistical models in a computer is a fundamental aspect of research in the field of nonparametric curve estimation. Methods such as the FFT (Fast Fourier Transform) or WARP (Weighted Average of Rounded Points) have been developed and analysed for computer implementation of the different techniques in this realm, with the aim of reducing the computation time as much as possible. In this work we analyse two techniques with this objective. These are the vectorization of the source code in which the different algorithms are implemented, and their distributed execution. It can be observed that the vectorization of the programs can improve the results obtained with techniques such as the FFT or WARP, or, in some cases, can prevent the use of these.     

Remaining childless in affluent economies: a comparison of France, West Germany, Italy and Spain, 1994–2001 Rester sans enfant dans des sociétés d'abondances: une comparaison de la France, l'Allemagne de l'Ouest et l'Espagne, 19994–2001

This article explores why women delay childbearing and increase their likelihood to remain childless in Spain, Italy, West Germany and France. We take a macro–micro perspective and show that national institutions influence women’s life transitions, in particular partnership and motherhood. For coupled women, we find two alternative modes out of childlessness. In countries with high direct and indirect child costs, like Spain and Italy, entering a male-breadwinner couple or occupying a stable and high-income position facilitates motherhood, while in the French context motherhood is most likely in a dual-earner partnership.      

Technology,Gender and Time: A Contribution to the Debate

Using the data collected from research carried out at two Portuguese universities, this article highlights the way university lecturers currently deal with space and time. Their professional activity is not externally subjected to a specific time geometry that defines their working day. Teaching and researching time–spaces are basically task oriented. Furthermore, university lecturers are increasingly requested to perform tasks in very short time periods in order to secure financial support. To a certain extent, this pressure upon their work is facilitated by the use of technology. However, technology also causes a profound reconfiguration of working times and spaces, especially because it allows lecturers to work at any time from anywhere. This not only renders university physical spaces and times virtual, but it also leads to the colonization of several other time–spaces of their social lives. This article analyses the ambivalence and difficulty of managing time–spaces, as described by the lecturers themselves, with the objective of providing a more accurate awareness of the implications of technology on their lives, as well as the hazards of the progressive externalization of academic work, particularly for women.