- Stochastic processes and statistical mechanics
- Bayesian Methods and Mixture Models
- Stochastic processes and financial applications
- Markov Chains and Monte Carlo Methods
- Statistical Methods in Clinical Trials
- Theoretical and Computational Physics
- Statistical Methods and Inference
- Advanced Statistical Methods and Models
- Benford’s Law and Fraud Detection
- Probability and Risk Models
- Bayesian Modeling and Causal Inference
- Matrix Theory and Algorithms
- Diffusion and Search Dynamics
- Probabilistic and Robust Engineering Design
- Digital Media Forensic Detection
- Nuclear reactor physics and engineering
- Mathematics, Computing, and Information Processing
- Probability and Statistical Research
- Mathematical Dynamics and Fractals
- Optimal Experimental Design Methods
- Mathematical functions and polynomials
- Evolution and Genetic Dynamics
- Multi-Criteria Decision Making
- Artificial Intelligence in Games
- Statistical and numerical algorithms
McMaster University
2009-2019
Max Planck Society
2014
Max Planck Institute for Plasma Physics
2013-2014
University of Michigan–Ann Arbor
1978-1987
Cornell University
1980
University of Alberta
1976-1978
University of Twente
1976
Princeton University
1975
We address the problem of predicting high detail RNA structure geometry from information available in low resolution experimental maps electron density. Here refers to ≥2.5Å where location phosphate groups and glyocosidic bonds can be determined density but all other backbone atom positions cannot. In contrast, determines atomic positions. To this end, we firstly create a gold standard data base for four manually corrected suites, each reflecting one out sugar pucker-pair configurations....
For the design and operation of large fusion devices a detailed understanding power exhaust processes is necessary.This paper will give an overview current research on divertor load mechanisms.The results shown are obtained in JET with ITER like wall ASDEX Upgrade tungsten coated plasma facing components.The challenges infrared thermography bulk presented.For steady state heat load, fall-off length λ q JET-ILW compared to empirical scaling found carbon components.A first attempt scale...
We show that there always exists a sequence of normalizing constants for the supercritical multitype Galton-Watson process so normalized converges in probability to limit which is proper and not identically zero. The Laplace-Stieltjes transform random variable characterized as unique solution under certain conditions vector Poincare functional equation.
Large deviation principles are established for some random combinatorial structures including the Ewens sampling formula and Pitman formula. A path-level large principle is former on cadlag space D$(o, 1], R)$ equipped with uniform convergence topology, rate function same as a Poisson process justifying approximation at level. total number of parts in partition obtained formula; here depends only one two parameters which display different roles scales. In addition to these results, we also...
A characteristic property of the Ewen sampling formula is shown to follow from invariance under size-biased Poisson–Dirichlet distribution.
A characteristic property of the Ewen sampling formula is shown to follow from invariance under size-biased Poisson–Dirichlet distribution.
Aims to provide in-depth descriptions of the latest developments in multiple comparison methods and selection procedures, while emphasizing biometry. This text is published honour 70th birthday Charles W. Dunnett - a pioneer statistical methodology.
Sj,, = n P(Ai,Ai ...? Ai.). (1.2) 1 1) of the A1,..., A, as being degree d. It is already clear from (1.1) and that Ar+i,n invariant under permutation labels A1,... , A,. We obtain bounds for Ar+,n with this same property, sharpening development in Seneta (1988). The are not terms 'aggregated' probabilities type Sj,,, but rather 'microprobabilities' P(AjA,2... A,). While a great deal investigation, often on optimality, beginning Bonferroni's original bounds, P(U Ai) {Sj,,} alone, has been...
The multitype Galton-Watson process is considered both with and without immigration. Proofs are given for the existence of invariant measures their uniqueness examined by functional equation methods. Theorem 2.1 proves uniqueness, under certain conditions, solutions a multidimensional Schröder equation. Regular variation shown to play role in theory.
This paper presents a best estimate plus uncertainty (BEPU) methodology applied to dryout, or critical channel power (CCP), modeling based on Monte Carlo approach. method involves the identification of sources and development error models for characterization separation epistemic aleatory uncertainties associated with CCP parameter. Furthermore, proposed facilitates use actual operational data leading improvements over traditional methods, such as sensitivity analysis, which assume...
The response of a complex physical system is often evaluated by tolerance interval for percentile the distribution variable interest that estimated best-estimate codes. This used as test statistic to make decisions about behavior in context specific safety issue. most common methods determine such intervals are order statistics leading so-called "95/95" level standard. application these predicated on assumption simulated responses identical those an actual reactor under postulated...
Résumé L'identité de Gumbel établit l'égalité la somme Bonferroni S k,n , k = 0, 1, 2, … n et du moment binomial d'ordre variable qui compte, dans un ensemble arbitraire événements, le nombre M d'événements se réalisant. Nous présentons traitement unifié bornes bien connues obtenues ce contexte par Bonferroni, Galambos‐Rényi, Dawson‐Sankoff Chung‐Erdös, ainsi que quelques moins établies Fréchet Gumbel. Toutes font intervenir des sommes Bonferroni. Notre démarche consiste à montrer ces...