摘要:

AbstractWhen the object of an experiment is to demonstrate equivalence of two treatments then, under normal theory, there are two competing methods: The doublet‐test and the test of Anderson and Hauck. Viewing the operating characteristic as a function of the noncentrality parameter the doublet‐test has slightly more power, especially for few degrees of freedom, in the upper range of the curve (power ≥0.6). In contrast, the test of Anderson‐Hauck is considerably more powerful towards the lower end of the powe

ISSN:0323-3847    

DOI:10.1002/bimj.4710320302

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractThe problem of comparing two treatments with respect to a continuous response variable, following an exponential distribution, is considered. It is assumed that a relevant comparative clinical trial is planned to either (i) test that the effect of one treatment is larger or not much smaller than that of the other, (ii) test whether the treatments are equivalent, or (iii) select the better treatment when they may not be equivalent and otherwise assert equivalence. Statistical procedures for achieving these goals are proposed and description of the required sample sizes is given with the necessary tables.

ISSN:0323-3847    

DOI:10.1002/bimj.4710320303

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractThe authors (1968) have previously given tables of the percentage points of the range (=w) ofrsamples from a multinomial distribution ofrcells each with probabilityr−1,r= 2(1)10. This paper presents the small sample power ofwforr= 2(1)6 under alternativesHto the null hypothesisH0of a symmetric multinomial distribution. The alternatives are of the formH: pi=r−1+c1n1/2i= 1, …,r, where thec's are such that ∑ci=0, i. e.Hconsists of a set of multinomial probabilities approachingH0of 0(n−1/2). Table 1 gives values of the power of the range test, and Table 2 consists of selected comparisons of the power with respect to theX2index of d

ISSN:0323-3847    

DOI:10.1002/bimj.4710320304

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractOften the statistician is faced with the necessity of analysinga × b‐contingency tables which are partially very sparsely occupied beyond a very‐well occupied “kernel”. The weak part shall not be due to misclassification or miscalculation. By the aid of a fictive example an analysing strategy is discussed combining exploratory and confirmatory elements and comparing the use of Exact Fisher Test and Freeman‐Halton Test with Pearson'sX2and Fisher's Maximum Likelihood Quotient St

ISSN:0323-3847    

DOI:10.1002/bimj.4710320305

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

ISSN:0323-3847    

DOI:10.1002/bimj.4710320306

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractThis paper suggests a method for explorative data analysis of samples of figures described byn‐tuples of landmarks. It uses some variants of a form of procrustes analysis introduced by D. G. Kendall. Their results can be analysed by well‐known methods of data analysis, for example by cluster analysis. As an example a sample of hands of students is stud

ISSN:0323-3847    

DOI:10.1002/bimj.4710320307

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

ISSN:0323-3847    

DOI:10.1002/bimj.4710320309

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractMethods usually applied for analysis of biological signals, as a rule, presuppose stationarity of the time series under investigation. Such kind of approach, however, will certainly be connected with an essential loss of information as dynamic properties of the system are excluded from consideration a priori. The method of time‐variant spectral analysis of the EEG suggested in this paper, on the contrary, provides time‐relatedchanges of the spectral coefficients. Additinonally, a decomposition of the EEG into its natural componentsas well as a combined EEG‐/EP‐analysis is

ISSN:0323-3847    

DOI:10.1002/bimj.4710320310

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

ISSN:0323-3847    

DOI:10.1002/bimj.4710320311

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY

摘要:

AbstractA Monte Carlo procedure is suggested for the problem of judging whether two sets of grouped circular data, possibly differently oriented, have arisen from otherwise equivalent distributions. The procedure is applied to data relating to the months of onset of two illnesses in a group of patients.

ISSN:0323-3847    

DOI:10.1002/bimj.4710320312

出版商:WILEY‐VCH Verlag    

年代:1990

数据来源: WILEY