Jean-Pierre Gauchi

Quantitative dietary exposure assessment of the Catalonian population (Spain) to the mycotoxin deoxynivalenol

The mycotoxin deoxynivalenol (DON) is one of the most common contaminants of cereals worldwide, and its occurrence has been widely reported in raw foods and foodstuffs, around the European region, including Catalonia. In the present work, a stochastic methodology has been applied to accurately assess the exposure of the Catalonian population (Spain) to DON through food consumption. Raw contamination data was provided by a large survey conducted in this region, in addition to the raw consumption data from a nutritional study specifically designed to assess the dietary intake of the main foodstuffs related to DON contamination for all population age groups. Contamination and consumption data were combined by simulation using an essentially parametric (P-P) method. The P-P method draws sampling values from distribution functions fitted to consumption and contamination data sets. Moreover, to quantify the accuracy and reliability of the statistics estimates, we built the related confidence intervals using a pseudo-parametric bootstrap method. Considering the results drawn from the P-P simulation method, the Catalonian population should be expected to be exposed at moderated levels of deoxynivalenol, the infants and individuals with ethnic dietary patterns being the most exposed population groups

Nonparametric particle filtering approaches for identification and inference in nonlinear state-space dynamic systems

Most system identification approaches and statistical inference methods rely on the availability of the analytic knowledge of the probability distribution function of the system output variables. In the case of dynamic systems modelled by hidden Markov chains or stochastic nonlinear state-space models, these distributions as well as that of the state variables themselves, can be unknown or untractable. In that situation, the usual particle Monte Carlo filters for system identification or likelihood-based inference and model selection methods have to rely, whenever possible, on some hazardous approximations and are often at risk. This review shows how a recent nonparametric particle filtering approach can be efficiently used in that context, not only for consistent filtering of these systems but also to restore these statistical inference methods, allowing, for example, consistent particle estimation of Bayes factors or the generalisation of model parameter change detection sequential tests.Real-life applications of these particle approaches to a microbiological growth model are proposed as illustrations.

New prediction interval and band in the nonlinear regression model: application to predictive modeling in foods.

This article is concerned with the proposal of a new prediction interval and band for the nonlinear regression model. The construction principle of this interval and band is based on an exact (the meaning of the term “exact” will be given later) confidence region for parameters of the nonlinear regression model. This region, fully described in Vila and Gauchi (2007), provides a rigorous justification for the new prediction interval and band that we propose. This new band is then compared to the classical bands (which are asymptotic and thus approximate for small n), and also to the band based on the bootstrap resampling method. The comparison of these bands is undertaken with simulated and real data from predictive modeling in food science.

Small Size Designs in Nonlinear Models Computed by Stochastic Optimization

Optimality criteria, that are functions of the mean square error matrix, are expressed as integrals of the density of the parameter estimator. The optimum design is obtained by an accelerated stochastic optimization method. The estimator is modified to reflect prior knowledge about the parameters, and to take into account the boundary of the parameter space. Results of (1992) are extended and improved by that. Computer results are presented on examples.