摘要:

A number of statistical tests are proposed for testing whether an observed covariance matrix is consistent with an expected covariance matrix deduced from an assumed composition of the variates. This aspect of multivariate analysis has no univariate analogue and as yet has not received proper treatment in the statistical literature. The general method proposed has been called ‘structural analysis’ and its role in multivariate hypothesis‐testing is discussed.Two different types of patterned covariance matrices arising out of various linear models are considered. The ‘reducible pattern’, composed of covariance matrices which can be diagonalized by pre‐ and post‐multiplication by a known matrix, and covariance matrices which can be transformed to a tridiagonal form (such as the Guttman quasi‐simplex, and the ‘aristocratic’ matrices), are discussed and illustrated with empirical data. Likelihood ratio tests for both classes of patterns have been constructed, and in each case the estimation procedure ha

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00438.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

Various statistical models for simplex structures are formulated in terms of the well‐known Wiener and Markov stochastic processes. A distinction is made between a perfect simplex and a quasi‐simplex. For each model the problems of identification and estimation of the parameters and that of testing the goodness of fit of the model are considered. All models may be estimated by a general method for covariance structures developed by Jöreskog (1970), but in some cases simpler methods may be used, in which case these are prese

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00439.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

This paper investigates several methods available for testing comparisons between treatments in a repeated‐measurement design, within the framework of the general multinormal model. Various ratios are considered, and the weakest assumptions required for the validity of eachFratio are presented. When severalFratios are valid, power considerations are introduced for purposes of choice. Methods enabling the testing of validity assumptions are also presented, together with alternative multivariate statistics which can be used when noFratio is valid. All procedures presented in the paper are illustrated by a numerical exampl

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00440.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

The discriminant function is presented in unnormalized and in normalized form. For the two‐variable case, certain parameter‐dependent features are discussed in Section 2, and sampling features (expectations and variances of various statistics) in Section 3. Section 4 comments on the usefulness of the discriminant function for classificat

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00441.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

Attention is drawn to the advantages, from the point of view of interpretation, of analysing multiway contingency tables by fitting constants to the marginal frequencies. It is then shown how the constants may be fitted by the method of maximum likelihood and estimates of their variances obtained. The procedure is illustrated by a worked example.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00442.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

A method of computer simulation for comparing the relative efficiencies of alternative experimental designs is applied to the study of the efficiency with which broad heritability can be estimated and the two main sources of environmental variation separated in data from human populations. Two minimal sets of data are compared with respect to these criteria and the effects of varying the proportions of the groups of relatives comprising a set are investigated. The relationship between efficiency, the level of broad heritability and the degree of dominance is noted, and the results related to those of a previous study.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00443.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

A model is developed for the case where a subject is presented withNalternatives from which he has to choose the single correct one. It is assumed that the subject randomly selects an alternative from a subset consisting of from 1 toNof the alternatives presented. For a given set of alternatives the size,i, of this choice set will depend on the subject's knowledge, which consequently may be measured by some suitable function ofi, or, alternatively, of (N — i). Two measures are considered, which require a modification of the usual test procedure, so that subjects are allowed repeated attempts until the correct alternative is selected. The model is shown to be successful in predicting performance in then‐alternative case fromN‐altentative performance

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00444.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

摘要:

A model for brightness contrast is presented which assumes two processes: an initial ‘perceptual’ transformation of visual intensity information, followed by lateral inhibitory interactions which can be represented linearly. The model has been applied to appropriate data and it appears to provide a satisfactory account of both brightness enhancement and brightness depression for homogeneous areas under stable conditions of viewing. This application of the model also provides evidence on the form of the function relating the relevant perceptual correlate of stimulus intensity to the physical values. This appears to be consistent with previous evidence that this relation is a power function with an exponent lying within a range about un

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00445.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1970.tb00446.x

出版商:Blackwell Publishing Ltd    

年代:1970

数据来源: WILEY