摘要:

AbstractThe paper deals with the statistical estimation of mean triangles of landmark data. For the model introduced by Bookstein (1986) three methods of estimating the “ideal” triangle are compared: the maximum likelihood method based on the exact distribution given in Mardia and Dryden (1989a), a moment method and an iterative algorithm yielding a mean triangle in the sense of Fréchet. These methods are compared by Monte Carlo simulation applied also to models with variances greater than those required for Bookstein's normal approxima

ISSN:0323-3847    

DOI:10.1002/bimj.4710350602

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractIn soil micromorphology fissures are considered in vertical sections. To get information about the properties of the soil the joint distribution of spatial direction and width is of interest. The fissures are mathematically generalized to flat bodies which form a stationary weighted surface process with the weight “thickness”. Because of stationarity a joint distribution of spatial direction and thickness exists in a “typical point” of the surface process. A suitable parametric family of distributions is assumed. The corresponding parameters can be estimated from measurements on the vertical sections. But on the sections only the visible thickness and the visible angle of a fissure can be measured. Therefore the joint distribution of these variables is expressed by the joint spatial distribution of spatial direction and thickness. This derived distribution depends on the same parameters. The Chi‐Square method is proposed for the parameter estimation. The estimation procedure is demonstrated using the Bingham‐Mardia distribution for the direction and the lognormal distribution for the thickness and by defining a way to correlate the mean thickness and th

ISSN:0323-3847    

DOI:10.1002/bimj.4710350603

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractIn soil micromorphology fissures are considered in vertical sections. To get information about the properties of the soil the joint distribution of spatial direction and width of these fissures is of interest. The fissures are mathematically generalized to flat bodies which are defined as stationary weighted surface processes with the weight “thickness”. In a typical point of the surface process suitable, joint parametric distributions of direction and thickness are assumed. The parameters have to be estimated from measurements on vertical sections which are taken from the soil. On these sections only a visible thickness and a visible angle can be observed. The joint distribution of these variables can be expressed by the joint distribution of spatial direction and thickness with the same parameters and in this indirect way the parameters can be estimated. The paper describes how to randomize the vertical section and how to measure the visible variables on the sections. The Chi‐Square method is proposed for the parameter estimation. Further it is discussed how to derive good starting values for the numerical procedure. All this is demonstrated in a simulation study using the Bingham‐Mardia distribution for the direction and the lognormal distribution for the thickness including a way to correlate the mean thickness and the direction. Finally an application in soil micromorphology is demonstrated for one soil

ISSN:0323-3847    

DOI:10.1002/bimj.4710350604

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractCalculating the required sample size for a desired power at a given type I error level, we often assume that we know the exact time of all subject responses whenever they occur during our study period. It is very common, however, in practice that we only monitor subjects periodically and, therefore, we know only whether responses occur or not during an interval. This paper includes a quantitative discussion of the effect resulting from data grouping or interval censoring on the required sample size when we have two treatment groups. Furthermore, with the goal of exploring the optimum in the number of subjects, the number of examinations per subject for test responses, and the total length of a study time period, this paper also provides a general guideline about how to determine these to minimize the total cost of a study for a desired power at a given α‐level. A specified linear cost function that incorporates the costs of obtaining subjects, periodic examinations for test responses of subjects, and the total length of a study period, is assumed, primarily for illustrative purpo

ISSN:0323-3847    

DOI:10.1002/bimj.4710350605

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractThis paper is concerned with comparing several increasing dose levels (treatments) with a zero dose control when the prior information about the umbrella pattern treatment means is available. The problem of testing whether there is at least one treatment which is better than the control is considered. Multiple test procedures are then proposed for deciding treatments (if any) which are better than the control. Some approximate criticial values of the proposed tests are reported. The results of a Monte Carlo power study are presented.

ISSN:0323-3847    

DOI:10.1002/bimj.4710350606

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractAn alternative way to implement the Akaike Information Criterion (AIC) is proposed for the evaluation of staging systems for colorectal cancer.

ISSN:0323-3847    

DOI:10.1002/bimj.4710350607

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractMonte Carlo Investigations have been widely used in Sample Surveys in Comparing the efficiency of various methods when exact mathematical comparisons are not possible. In this paper the same has been used for comparing the efficiency of Stratified Random Sampling with respect to Simple Random Sampling for estimation of Relative Risk in Case‐Control Studies. The data used relate to a Case Control study on peptic ulcer. On the basis of Monte Carlo Investigations on 50 samples of size 10–20 (Cases and Controls), it has been observed that there is considerable gain in efficiency in using Stratified Random Sampling over Simple Random Sampling. The sensitivity of the results with the change in Sample Size has also been investiga

ISSN:0323-3847    

DOI:10.1002/bimj.4710350608

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractIt is known that when the logistic response function is generalized by introducing shape parameters, the usual computational simplicity afforded by statistical softwares such as GLIM and S‐Plus may be lost. This fact is illustrated by Prentice (1976), Brown (1982), Stukel (1985, 1988), and El‐Saidi (1986). In this paper, we consider a power transformation of the generalized logistic model and show how the use of such transformation simplifies the computational difficulties associated with generalized logistic models. Furthermore, applying this technique to some data sets previously analyzed by D'Angio et al. (1981) and Brown (1982) shows an improvement in the fit in comparison to other models such as the logistic, the unstratified multiplicative model GMU and the additive model GA described by Storer et al. (19

ISSN:0323-3847    

DOI:10.1002/bimj.4710350609

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractThis paper considers a modification of famous unrelated question randomized response model due to Moors' (1971). The estimator based on the proposed strategy is shown to be unbiased and always more efficient than the estimator for Moor's model.

ISSN:0323-3847    

DOI:10.1002/bimj.4710350610

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY

摘要:

AbstractIn this paper (1) expressions (correct ton−2terms) for biases, variances, and covariances of the estimatorsaandbof Hermite distribution with probability generating function Exp[a(t–1) +b(t–1)] are obtained for two mixed moment estimates; (2) for the biases and variance‐covariances, approximate regions of the parameter space (a>0,b>0) have been outlined where a sample of size 100 can be considered as “safe” in the sense that contribution of second order terms in them is 5% of that from the first order term; (3) comparison of the biases and variance‐covariances of these two sets of estimators are made with those for the moment estimators, maximum likelihood estimates and the even point estimators for a sample of size 100 using the terms up to ordern−2; (4) the comparisons based onn−2terms in (3) have not only provided information on the estimation procedures included in the Hermite distribution, but also demonstrated the importance of higher order terms in the sampling properties of the various alternative techniques for the Her

ISSN:0323-3847    

DOI:10.1002/bimj.4710350611

出版商:WILEY‐VCH Verlag    

年代:1993

数据来源: WILEY