Search results for "Statistical"

Reference Posterior Distributions for Bayesian Inference

1979

Bayesian measures of surprise for outlier detection

2003

From a Bayesian point of view, testing whether an observation is an outlier is usually reduced to a testing problem concerning a parameter of a contaminating distribution. This requires elicitation of both (i) the contaminating distribution that generates the outlier and (ii) prior distributions on its parameters. However, very little information is typically available about how the possible outlier could have been generated. Thus easy, preliminary checks in which these assessments can often be avoided may prove useful. Several such measures of surprise are derived for outlier detection in normal models. Results are applied to several examples. Default Bayes factors, where the contaminating…

New adaptive synchronization algorithm for a general class of complex hyperchaotic systems with unknown parameters and its application to secure comm…

2022

Abstract The aim of this report is to investigate an adaptive synchronization (AS) for the general class of complex hyperchaotic models with unknown parameters and a new algorithm to achieve this type of synchronization is proposed. Owing to the intricacy behavior of hyperchaotic models that could be effective in secure communications, the special control based on adaptive laws of parameters is constructed analytically, and the corresponding simulated results are performed to validate the algorithm’s accuracy. The complex Rabinovich model is utilized as an enticing example to examine the proposed synchronization technique. A strategy for secure communication improving the overall cryptosyst…

What subject matter questions motivate the use of machine learning approaches compared to statistical models for probability prediction?

2014

This is a discussion of the following papers: "Probability estimation with machine learning methods for dichotomous and multicategory outcome: Theory" by Jochen Kruppa, Yufeng Liu, Gerard Biau, Michael Kohler, Inke R. Konig, James D. Malley, and Andreas Ziegler; and "Probability estimation with machine learning methods for dichotomous and multicategory outcome: Applications" by Jochen Kruppa, Yufeng Liu, Hans-Christian Diener, Theresa Holste, Christian Weimar, Inke R. Konig, and Andreas Ziegler.

Monte Carlo simulation of polymers at interfaces

1993

Abstract Polymers at interfaces pose challenging problems to statistical physics because their configurations often differ greatly from the bulk. Computer simulation of coarse-grained models then gives valuable insight and allows stringent tests of various theoretical predictions. Three examples are briefly treated: chain configurations of B-chains in the surface-enriched B-rich layer of an (AB) binary polymer mixture; “frustrated” lamellar ordering in ultra-thin block-copolymer films; and the collapse of polymer brushes in bad solvents.

Cutting rules and positivity in finite temperature many-body theory

2022

Abstract For a given diagrammatic approximation in many-body perturbation theory it is not guaranteed that positive observables, such as the density or the spectral function, retain their positivity. For zero-temperature systems we developed a method [2014 Phys. Rev. B 90 115134] based on so-called cutting rules for Feynman diagrams that enforces these properties diagrammatically, thus solving the problem of negative spectral densities observed for various vertex approximations. In this work we extend this method to systems at finite temperature by formulating the cutting rules in terms of retarded N-point functions, thereby simplifying earlier approaches and simultaneously solving the issu…

On powerful exact nonrandomized tests for the Poisson two-sample setting.

2020

In the case of two independent samples from Poisson distributions, the natural target parameter for hypothesis testing is the ratio of the two population means. The conditional tests which have been derived for this class of problems already in the 1940s are well known to be optimal in terms of power only when randomized decisions between hypotheses are admitted at the boundary of the respective rejection regions. The major objective of this contribution is to show how the approach used by Boschloo in 1970 for constructing a powerful nonrandomized version of Fisher’s exact test for hypotheses about the odds ratio between two binomial parameters can successfully be adapted for the Poisson c…

Role of the noise on the transient dynamics of an ecosystem of interacting species

2002

Abstract We analyze the transient dynamics of an ecosystem described by generalized Lotka–Volterra equations in the presence of a multiplicative noise and a random interaction parameter between the species. We consider specifically three cases: (i) two competing species, (ii) three interacting species (one predator–two preys), (iii) n-interacting species. The interaction parameter in case (i) is a stochastic process which obeys a stochastic differential equation. We find noise delayed extinction of one of two species, which is akin to the noise-enhanced stability phenomenon. Other two noise-induced effects found are temporal oscillations and spatial patterns of the two competing species. In…

Role of sub- and super-Poisson noise sources in population dynamics

2020

In this paper we present a study on pulse noise sources characterized by sub- and super-Poisson statistics. We make a comparison with their uncorrelated counterpart. i.e. pulse noise with Poisson statistics, while showing that the correlation properties of sub- and super-Poisson noise sources can be efficiently applied to population dynamics. Specifically, we consider a termite population, described by a Langevin equation in the presence of a pulse noise source, and we study its dynamics and stability properties for two models. The first one describes a population of several colonies in a new territory with adverse environmental conditions. The second one considers the development of a sing…

Deducing self-interaction in eye movement data using sequential spatial point processes

2016

Eye movement data are outputs of an analyser tracking the gaze when a person is inspecting a scene. These kind of data are of increasing importance in scientific research as well as in applications, e.g. in marketing and man-machine interface planning. Thus the new areas of application call for advanced analysis tools. Our research objective is to suggest statistical modelling of eye movement sequences using sequential spatial point processes, which decomposes the variation in data into structural components having interpretation. We consider three elements of an eye movement sequence: heterogeneity of the target space, contextuality between subsequent movements, and time-dependent behaviou…