摘要:

AbstractThis paper suggests some characteristics which may help to detect and quantify “correlations” between planar geometrical structures such as point patterns/point processes, fibre systems/fibre processes and region patterns/random sets. Typical examples of “correlations” are mutual attraction or inhibition. Some of the characteristics are related to moment‐measures, the definition of the others is similar to that of quantities in the mathematical morphology of MATHERONand SERRA. For all quantities unbiased estimators are given. Also the detection of orientations in random structures is discussed. An ecological example demonstrates the application of the charac

ISSN:0323-3847    

DOI:10.1002/bimj.4710240702

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

ISSN:0323-3847    

DOI:10.1002/bimj.4710240703

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractWe consider the one‐way classification random model. Using a measure of unbalancedness some MINQ‐Estimation procedures and ANOVA‐procedure have been investigated with respect to their sensibility to unbalancedness.Further if we define the efficiency of a designNas the ratio of the MSE (.)'s of the balanced and the unbalanced case we obtain for three MINQE‐procedures upper and lower bounds of the minimum eff

ISSN:0323-3847    

DOI:10.1002/bimj.4710240704

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

ISSN:0323-3847    

DOI:10.1002/bimj.4710240705

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractThe concept of balanced sampling is applied to prediction in finite samples using model based inference procedures. Necessary and sufficient conditions are derived for a general linear model with arbitrary covariance structure to yield the expansion estimator as the best linear unbiased predictor for the mean. The analysis is extended to produce a robust estimator for the mean squared error under balanced sampling and the results are discussed in the context of statistical genetics where appropriate sampling produces simple efficient and robust genetic predictors free from unnecessary genetic assumptions.

ISSN:0323-3847    

DOI:10.1002/bimj.4710240706

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractIn the present paper, a generalized cosine formula is obtained for the difference between two angles in ak‐dimensional Euclidean space, which is utilized to derive one of the principles of path analysis, namely, the correlation between two random variables is the sum of all connecting paths between the

ISSN:0323-3847    

DOI:10.1002/bimj.4710240707

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractA generalized algorithm is presented for generating the ensemble of uniquer×ccontingency tables with arbitrary, fixed row and column sums, and of arbitrary, even large, sizes. This further contributes to the operational solution of the exact ensemble distribution of 2‐way contingency tables and cross classifications, along with the sampling distribution of the conventional χ2‐like test statistic, or of any other test statistic calculated from the tables. The associated enumeration problem of Integer arrays with fixed row and column sums has been operationally s

ISSN:0323-3847    

DOI:10.1002/bimj.4710240708

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractIn this paper the analysis of covariance in the split block design with many concomitant variables is presented. The problems concerning the estimation of parametric functions and testing hypotheses are discussed. In the presentation of the model three kinds of regression coefficients for individual sources of variation are taken into consideration. It is shown that for every estimable function of fixed effects, the best linear unbiased estimator under the assumed model is the same as the best linear unbiased estimator under the model with covariance matrix equal to identity matrix multiplied by a positive constant. A variance of this estimator can be calculated by the method presented here. Test functions for standard hypotheses concerning fixed effects are obtained.

ISSN:0323-3847    

DOI:10.1002/bimj.4710240709

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractFor testing a finite number of linear hypotheses the SANOVA method can be used. This method is based on the multivariateFdistribution. In special case the multivariateFdistribution depends on correlation coefficients of so called “accompanying” normal distribution. In this paper, formulas for calculating these coefficients are given. Some experimental designs are discussed in more detail. Theoretical considerations are illustrated by numerical examp

ISSN:0323-3847    

DOI:10.1002/bimj.4710240710

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY

摘要:

AbstractIn this article we give a simple procedure to determine the exact distribution of the likelihood ratio test of a statistical hypothesis regarding the parameter of the uniform distribution. The resulting distribution will be shown to serve as an approximation to the distribution of the likelihood ratio statistic for testing the equality of scale parameters ofkindependent Exponential populations.

ISSN:0323-3847    

DOI:10.1002/bimj.4710240711

出版商:WILEY‐VCH Verlag    

年代:1982

数据来源: WILEY