A5071251463- Statistical Methods and Inference
- Advanced Statistical Methods and Models
- Statistical Distribution Estimation and Applications
- Probability and Risk Models
- Bayesian Methods and Mixture Models
- Probabilistic and Robust Engineering Design
- Mathematical Approximation and Integration
- Control Systems and Identification
- Advanced Statistical Process Monitoring
- advanced mathematical theories
- Bayesian Modeling and Causal Inference
- Mathematical functions and polynomials
- Statistical Methods and Bayesian Inference
- Advanced Topology and Set Theory
- Numerical methods in inverse problems
- Advanced Banach Space Theory
- Computability, Logic, AI Algorithms
- Color Science and Applications
- Advanced Computational Techniques in Science and Engineering
- Markov Chains and Monte Carlo Methods
- Graph theory and applications
- Textile materials and evaluations
- Holomorphic and Operator Theory
- Advanced Queuing Theory Analysis
- Mathematical Inequalities and Applications
Norwegian University of Science and Technology
2011-2022
Institute of Microelectronics Technology and High Purity Materials
1995-2016
National and Kapodistrian University of Athens
2015
Russian Academy of Sciences
2001-2012
Lomonosov Moscow State University
2007
Abstract. Several old and new density estimators may have good theoretical performance, but are hampered by not being bona fide densities; they be negative in certain regions or integrate to 1. One can therefore simulate from them, for example. This paper develops general modification methods that turn any estimator into one which is a density, always better performance under set of conditions arbitrarily close complementary conditions. improvement‐for‐free procedure can, particular, applied...
Abstract In this paper, we suggest a new method of bandwidth selection in kernel density estimation. The selector is less subject to the undersmoothing effect than AMISE (asymptotic mean integrated square error) optimal bandwidth. Keywords: estimationbandwidth selectionnonparametric estimation Acknowledgements authors thank two referees for useful comments which improved presentation paper. This research was supported part by RFBR grant No. 11–01–00515a.
The uniqueness and stability conditions of reconstructing a distribution independent identically distributed random variables $X_1,\ldots,X_m$ by the sum $S=X_1+\cdots+X_m$ for fixed~m are given. This paper considers two generalizations problem variables~$X_j$: $S=\gamma_1X_1+\cdots+\gamma_mX_m$, where variables~$\gamma_j$ take values~0 and~1 with some fixed probabilities, bythe $S_N=X_1+\cdots+X_N$ number~N summands~$X_j$. In these problems there given not only sufficient but quantitative...
A new statistical method for estimating the orientation distribution of fibres in a fibre process is suggested where observed form degraded digital greyscale image. The based on line transect sampling image few fixed directions. well-known stereology available if intersections between transects and can be counted. We extend this to case where, instead intersection points, only scaled variograms grey levels along are observed. nonlinear estimation equations parametric as well numerical...
Previous article Next Some Inequalities for Characteristic Functions of Unimodal DistributionsN. G. UshakovN. Ushakovhttps://doi.org/10.1137/1126065PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAbout FiguresRelatedReferencesCited byDetails On an inequality by Kulikova and ProkhorovStatistics & Probability Letters, Vol. 79, No. 14 Cross Ref Uniform Distributions on Convex Sets: Inequality FunctionsA. A. Yu. V. Prokhorov25 July 2006 | Theory Its...
In this article, we consider the problem of nonparametric density estimation in case, when original sample has a large size, but data are given binned form, i.e. form histogram. Such situations typical for many physical problems, particular, scanning electron microscopy and beam lithography. We study how superkernels can be used such situations. It is shown that essentially superior over conventional kernels not only very smooth densities. The bandwidth bin width selection also considered.
In this work, non parametric tests are proposed for testing the homogeneity of two or more populations. The based on recently obtained characterizations. test procedure is permutation bootstrap technique. For two-sample case new compared with empirical characteristic function and some other tests. comparison fulfilled via a Monte Carlo simulation.