Displaying 21 – 40 of 100

Showing per page

Criteria for optimal design of small-sample experiments with correlated observations

Andrej Pázman (2007)

Kybernetika

We consider observations of a random process (or a random field), which is modeled by a nonlinear regression with a parametrized mean (or trend) and a parametrized covariance function. Optimality criteria for parameter estimation are to be based here on the mean square errors (MSE) of estimators. We mention briefly expressions obtained for very small samples via probability densities of estimators. Then we show that an approximation of MSE via Fisher information matrix is possible, even for small...

Design of a model following control system for nonlinear descriptor system in discrete time

Shujing Wu, Shigenori Okubo, Dazhong Wang (2008)

Kybernetika

A model following control system (MFCS) can output general signals following the desired ones. In this paper, a method of nonlinear MFCS will be extended to be a nonlinear descriptor system in discrete time. The nonlinear system studied in this paper has the property of norm constraint | | f ( v ( k ) ) | | α + β | | v ( k ) | | γ , where α 0 , β 0 , 0 γ < 1 . In this case, a new criterion is proposed to ensure the internal states be stable.

Diseño secuencial para discriminar entre modelos, basado en la información cuadrática.

M.ª Pilar García-Carrasco Aponte (1985)

Trabajos de Estadística e Investigación Operativa

La información cuadrática es una buena alternativa a la información de Shannon para todos aquellos problemas que, por su naturaleza, interesa tratarlos con una utilidad no local. El objetivo de este trabajo es dar, para estas situaciones, un método secuencial de construcción de diseños para discriminación entre modelos, basado en la maximización de la información cuadrática.Después de una introducción, donde se resumen los conceptos y resultados principales sobre información cuadrática, se plantea...

D-optimal and highly D-efficient designs with non-negatively correlated observations

Krystyna Katulska, Łukasz Smaga (2016)

Kybernetika

In this paper we consider D-optimal and highly D-efficient chemical balance weighing designs. The errors are assumed to be equally non-negatively correlated and to have equal variances. Some necessary and sufficient conditions under which a design is D*-optimal design (regular D-optimal design) are proved. It is also shown that in many cases D*-optimal design does not exist. In many of those cases the designs constructed by Masaro and Wong (2008) and some new designs are shown to be highly D-efficient....

D-optimal cyclic two-dimensional block designs.

Jagdish N. Srivastava, A. M. Wijetunga (1983)

Trabajos de Estadística e Investigación Operativa

In this paper we consider a class of incomplete block designs in which every block is two dimensional. Thus heterogeneity is removed not only between blocks, but also in two directions within each block. Such designs have been considered before in Srivastava (1977,1978). Here, we consider the class of cyclic designs of this type when the number of treatments ν is an odd number between 5 and 25, and present designs that are D-optimal within this class.

H 2 -optimal rejection with preview: geometric constraints and dynamic feedforward solutions via spectral factorization

Elena Zattoni (2008)

Kybernetika

In this work, a feedforward dynamic controller is devised in order to achieve H2-optimal rejection of signals known with finite preview, in discrete-time systems. The feedforward approach requires plant stability and, more generally, robustness with respect to parameter uncertainties. On standard assumptions, those properties can be guaranteed by output dynamic feedback, while dynamic feedforward is specifically aimed at taking advantage of the available preview of the signals to be rejected, in...

Information contained in design points of experiments with correlated observations

Andrej Pázman (2010)

Kybernetika

A random process (field) with given parametrized mean and covariance function is observed at a finite number of chosen design points. The information about its parameters is measured via the Fisher information matrix (for normally distributed observations) or using information functionals depending on that matrix. Conditions are stated, under which the contribution of one design point to this information is zero. Explicit expressions are obtained for the amount of information coming from a selected...

Currently displaying 21 – 40 of 100