 |
ECOLE SUPÉRIEURE D'ÉLECTRICITÉ Une grande école d'ingénieurs au cœur des sciences de l'information,de l'énergie et des systèmes |
 |
Journal articles
A Statistical Inference Comparison for Measurement Estimation using Stochastic Simulation Techniques
Abstract : The purpose of this paper is to present the comparison of different techniques for making statistical inference about a measurement system model. This comparison involves results when two main assumptions are made. First, the unknowable behavior of the probability density function (pdf) }(e) of errors, since the real measurement systems are always exposed to continuous perturbations of an unknown nature; second, the assumption that after some experimentation one can obtain sufficient information which can be incorporated into the modelling as prior information. The first assumption lead us to propose the use of two approaches which permit building hybrid algorithms; such approaches are the non-parametric bootstrap and the kernel methods. The second assumption makes possible the exploration of a Bayesian framework solution and the Monte Carlo Markov Chain (MCMC) auxiliary use to access the a posteriori pdf of the estimated measurand. For both assumptions over }(e) and the model, different classical criteria can be used; one uses also an extension of a recent criterion of entropy minimization. The entropy criterion is constructed on the basis of a symmetrized kernel estimate b}n;h(e) of }(e). Finally, a comparison between results obtained with the different proposed schemes is presented.
Cited literature [39 references]
https://hal-supelec.archives-ouvertes.fr/hal-00287379
Contributor : Karine El Rassi Connect in order to contact the contributor
Submitted on : Wednesday, June 11, 2008 - 5:11:50 PM
Last modification on : Wednesday, December 16, 2020 - 12:54:02 PM
Long-term archiving on: : Friday, September 28, 2012 - 3:51:14 PM
Contributor : Karine El Rassi Connect in order to contact the contributor
Submitted on : Wednesday, June 11, 2008 - 5:11:50 PM
Last modification on : Wednesday, December 16, 2020 - 12:54:02 PM
Long-term archiving on: : Friday, September 28, 2012 - 3:51:14 PM
File
IEEEart4.pdf
Files produced by the author(s)