Search results for "gradient"
showing 10 items of 725 documents
Fast Estimation of the Median Covariation Matrix with Application to Online Robust Principal Components Analysis
2017
International audience; The geometric median covariation matrix is a robust multivariate indicator of dispersion which can be extended without any difficulty to functional data. We define estimators, based on recursive algorithms, that can be simply updated at each new observation and are able to deal rapidly with large samples of high dimensional data without being obliged to store all the data in memory. Asymptotic convergence properties of the recursive algorithms are studied under weak conditions. The computation of the principal components can also be performed online and this approach can be useful for online outlier detection. A simulation study clearly shows that this robust indicat…
Recursive estimation of the conditional geometric median in Hilbert spaces
2012
International audience; A recursive estimator of the conditional geometric median in Hilbert spaces is studied. It is based on a stochastic gradient algorithm whose aim is to minimize a weighted L1 criterion and is consequently well adapted for robust online estimation. The weights are controlled by a kernel function and an associated bandwidth. Almost sure convergence and L2 rates of convergence are proved under general conditions on the conditional distribution as well as the sequence of descent steps of the algorithm and the sequence of bandwidths. Asymptotic normality is also proved for the averaged version of the algorithm with an optimal rate of convergence. A simulation study confirm…
Quantitative analysis of numerical estimates for the permeability of porous media from lattice-Boltzmann simulations
2010
During the last decade, lattice-Boltzmann (LB) simulations have been improved to become an efficient tool for determining the permeability of porous media samples. However, well known improvements of the original algorithm are often not implemented. These include for example multirelaxation time schemes or improved boundary conditions, as well as different possibilities to impose a pressure gradient. This paper shows that a significant difference of the calculated permeabilities can be found unless one uses a carefully selected setup. We present a detailed discussion of possible simulation setups and quantitative studies of the influence of simulation parameters. We illustrate our results b…
A gradient-based deletion diagnostic measure for generalized linear mixed models
2016
ABSTRACTA gradient-statistic-based diagnostic measure is developed in the context of the generalized linear mixed models. Its performance is assessed by some real examples and simulation studies, in terms of ability in detecting influential data structures and of concordance with the most used influence measures.
The “ThreePlusOne” Likelihood-Based Test Statistics: Unified Geometrical and Graphical Interpretations
2014
The presentation of the well known Likelihood Ratio, Wald and Score test statistics in textbooks appears to lack a unified graphical and geometrical interpretation. We present two simple graphical representations on a common scale for these three test statistics, and also the recently proposed Gradient test statistic. These unified graphical displays may favour better understanding of the geometrical meaning of the likelihood based statistics and provide useful insights into their connections.
Inferential tools in penalized logistic regression for small and sparse data: A comparative study.
2016
This paper focuses on inferential tools in the logistic regression model fitted by the Firth penalized likelihood. In this context, the Likelihood Ratio statistic is often reported to be the preferred choice as compared to the ‘traditional’ Wald statistic. In this work, we consider and discuss a wider range of test statistics, including the robust Wald, the Score, and the recently proposed Gradient statistic. We compare all these asymptotically equivalent statistics in terms of interval estimation and hypothesis testing via simulation experiments and analyses of two real datasets. We find out that the Likelihood Ratio statistic does not appear the best inferential device in the Firth penal…
The purification and properties of nucleoside phosphotransferase from mucosa of chicken intestine
1984
Abstract (1) Nucleoside phosphotransferase (nucleotide:3′-deoxynucleoside 5′-phosphotransferase, EC 2.7.1.77) has been purified from chicken intestine mucosa to apparent homogeneity. The enzyme is represented by a multisubunit protein at different degrees of association. It can dissociate into a compoenent with a marked fall in catalytic activity. (2) The associated forms are similar to the enzyme previously purified from chick embryo as regards: substrate specificity both with respect to nucleoside monophosphate donors and to deoxyribonucleoside acceptors; sigmoidicity in the rate curve with a variable phosphate donor; instability to heat, dilution and lowering of pH; the activating and pr…
Sticking Probability on Zeolites
2006
The sticking coefficient, i.e., the probability that, on hitting the surface of a nanoporous particle (zeolite), a molecule shall be able to enter the intracrystalline space, is a key quantity for the application of such materials in heterogeneous catalysis and molecular sieving. On the basis of pulsed field gradient NMR diffusion measurements and molecular dynamics simulations, typical values of this probability are found to be close to one. They exceed previous estimates on the basis of IR uptake measurements by many orders of magnitude.
Stochastic algorithms for robust statistics in high dimension
2016
This thesis focus on stochastic algorithms in high dimension as well as their application in robust statistics. In what follows, the expression high dimension may be used when the the size of the studied sample is large or when the variables we consider take values in high dimensional spaces (not necessarily finite). In order to analyze these kind of data, it can be interesting to consider algorithms which are fast, which do not need to store all the data, and which allow to update easily the estimates. In large sample of high dimensional data, outliers detection is often complicated. Nevertheless, these outliers, even if they are not many, can strongly disturb simple indicators like the me…
A symmetric BEM approach to strain gradient elasticity for 2D static boundary-value problems
2014
The symmetric Galerkin Boundary Element Method is used to address a class of strain gradient elastic materials featured by a free energy function of the (classical) strain and of its (first) gradient. With respect to the classical elasticity, additional response variables intervene, such as the normal derivative of the displacements on the boundary, and the work-coniugate double tractions. The fundamental solutions - featuring a fourth order partial differential equations (PDEs) system - exhibit singularities which in 2D may be of the order 4 1/ r . New techniques are developed, which allow the elimination of most of the latter singularities. The present paper has to be intended as a resear…