摘要:

We define generalized cuts the coefficients of which are congruent the coefficients of GOMARY-cuts. Congruent cuts are called related and belong to the same family of cuts. Using these ideas it will be possible to give conditions for sharpening a GOMORY cut to a extremal cut. Here an extremal cut is a cut going through a lattice point, which is optimal under the conditions defining the continuous optimum.

ISSN:0047-6277    

DOI:10.1080/02331887408801185

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

The model of genetic adaption is a stochastic search method for discrete non-linear optimization problems derived from biological evolution. Due to alternative mutations (i.e. small changes in the structures to be optimized) and selection (of the structure most suitable for this purpose) structurelines are developed. The results of expreience obtained during the optimization processes (statistical correlation between structures at a poor sub-optimal value is avoided by parallel search using several structures. The method of genetic adaptation is superior to the common blind search. The convegence of the genetic-adaptation method can be compared with that of the blind search. The convegence of the genetic-adaptation method can be compared with that of the blind seach by a series of practical examples.

ISSN:0047-6277    

DOI:10.1080/02331887408801186

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

In the present paper the dependence of a certain optimization problem on a parametertis studied. The problem may have the form min. As opposed to earlier papers the parameter.tin the present case runs through the wholeE1and not through the convexity intervalK0only. The main results are a notion of stability, which is suitable for such problems, and, associated with chis notion a partition theorem for the solvability region of this problem, which a t the same time gives the dependence of the optimal set and the solution function on the parametert

ISSN:0047-6277    

DOI:10.1080/02331887408801187

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

We consider two classes of risk functions and prove the nonadmissibility resp. admisibility of the usual estimatorfor the meanμof a normal distribution. Furthere, we consider the criteria of G- and D-optimality of experimental designs in connection with choice of a suitable estimator forμ

ISSN:0047-6277    

DOI:10.1080/02331887408801188

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

For a system of stochastic linear differnce equationswhere the random disturbancesUis a timely non-independent GAUssian process we derive BAYESian estimators forA,Band the unknown parameters the convariance function ofUtand conditions for the consistance this estimators. Further we derive BAYEsian predictors ofX(T+ 1) by the observation ofand the recursive form the BAYEsian estimators for time independent random disturbancesUt. For the derivation of the BAYEsian estimators and predictors conjugate prior distributions are used.

ISSN:0047-6277    

DOI:10.1080/02331887408801189

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

In addition to the article [1] a proof of the ixisieiice of optimal stratifications is presented. Further it is shown that the DALENIUS-equations are necessary conditions for the optimality of a stratification, even if the variance of the stratified sample mean is not everywhere differentiable. Finally an example of a density is given for which the solution procedure presented in [1] splits up into several branches. Notations and arrangement follows article [1].

ISSN:0047-6277    

DOI:10.1080/02331887408801191

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor

摘要:

This paper extends some well known distribution-free structural results in the univariate and multivariate hypothese testing to stochastic processes. The study of distribution-free tests is essentially the study of similar sets and test functions. It is found that (i) all distribution-free statistics are based on permuations through of the basic PITMAN functions; (ii) the optimal tests are based on the likelihoood function of the alternatives; (iii) that a generation of DARMOIS-PITMAN-KOOPMAN families. Furthermore, it is observed that for martingales and MARKOV processes asymptotically optimal tests can be constructed by employing various limit theorems.

ISSN:0047-6277    

DOI:10.1080/02331887408801192

出版商:Akademie-Verlag    

年代:1974

数据来源: Taylor