摘要:

A method is proposed for comparing the accuracy of diagnostic tests that rely on a reader's subjective interpretation of the results. An ANOVA approach is applied where the dependencies between the readers' estimates of diagnostic accuracy are handled by adjusting the usual F statistic for theestimated correlation. The distribution of the resulting test statistic is evaluated. The problem is particularly relevant to diagnostic radiology where multi-reader studies of the indices associated with the ROC curve serve important roles in evaluating the efficacy of diagnostic tests.

ISSN:0361-0918    

DOI:10.1080/03610919508813243

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

In this paper, methods are proposed in finding the robust design in both Taguchi and Standard setups when a signal factor is present. The robust design is a set of level combinations of control factors so that the effect of controllable noise factors on response is minimum. Both univariate and multivariate methods are used in finding the influential noise factors for the determination of robust designs.

ISSN:0361-0918    

DOI:10.1080/03610919508813244

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

An approximation to the p-values of the standard Cochran's and Hartley's tests for testing the homogeneity of variances is obtained for the case when the observations are autocorrelated. More specifically, the expression for calculating the p-value of Cochran's test for k(≥2) populations is given in terms of a multiple integral, whereas a series form is presented for the case of Hartley'stest. As, in practice, the tests for homogeneity of variances for 2 or 3 populations are often encountered, the closed form expressions for the p-values of these two cases are provided. For Cochran's test, we also provide suitable upper and lower bounds of the p-value for the general case of kpopulations. Numerical discussions are also given.

ISSN:0361-0918    

DOI:10.1080/03610919508813245

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

It is shown that the joint maximum likelihood estimator of slope and intercept of the regression line in the classical (known error-variance ratio) linear structural relationship model can be represented as a solution of a two- dimensional M- equation. Therefore, it is possible to use a general saddlepoint approximation for multidimensional M- equations. Under normality assumptions we express the solution of the implicit multivariate “centering equation” in an explicit form. This allows a considerable saving of computing time. By integrating out numerically an unwanted variable one is also able to find the saddlepoint approximation for the slope- estimator. Numerical examples illustrate the efficiency of the approximation.

ISSN:0361-0918    

DOI:10.1080/03610919508813246

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

The aim of this article is to construct a distribution–free test based on the method of weighted rankings for testing the equality of treatment effects in a randomized block design which, in terms of efficiency, is superior to both the Quade and Friedman tests under normality, being almost as efficient as the variance-ratio test, but whose performance remains quite reasonable under nonnormality. A conjecture which, according to Tardif (1987, JASA 82,,637-644), should ameliorate the performance of the Quade test relative to the Friedman test is first examined but it is seen that it does not provide a satisfactory solution. On the other hand, a maximization of the asymptotic efficiency for a fixed set of within-block scores is considered and, by allowing the between-block scores to depend on the number of observations per block, it gives rise to the formulation of an optimal test outperforming both the Quade and Friedman tests under normality. A Monte Carlo investigation finally examines the performance of this optimal test in small samples under normal, Laplace or Cauchy errors. Based on this study, its Fisher approximation is recommended when the number of observations per block takes small to moderate values

ISSN:0361-0918    

DOI:10.1080/03610919508813247

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

This paper focuses on the development of an ecological regression approach for voter transition estimation, avoiding the arbitrary assumptions in Goodman's classical model of ecological regression (Goodman [1959]). In doing this, we further develop previous attempts made at the ridge regression approach, by applying a modified generalized ridge estimator. To illustrate the application of the proposed estimation technique, we estimate the transition probabilities in the last two Swedish elections on a local level.

ISSN:0361-0918    

DOI:10.1080/03610919508813248

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

In this paper we consider a model based approach for estimating hierarchical kappa statistics to demonstrate improved precision through modelling. An extensive simulation study shows the improvement over the classical estimates due to model smoothing in terms of reduced mean square error

ISSN:0361-0918    

DOI:10.1080/03610919508813249

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

The Shewhart R–and S–charts are often used to monitor the variability of a quality characteristic of interest. In order to improve the sensitivity of these charting procedures to detecting small shifts in the process standard deviation, runs rules have been suggested. We evaluate the properties of the run length distributions of these charting procedures using a Markov chain approach. We show that the rules commonly used in practice based on the Western Electric Handbook do not have the statistical performance one might expect. Alternative runs rules are provided that provide the same in-control ARL's as the usual X-chart with runs rules. Average run length comparisons are made among various charts for monitoring variability, including cumulative sum charts and exponentially weighted moving average charts. It is shown that detection of decreases in variability is problematic with all of these methods.

ISSN:0361-0918    

DOI:10.1080/03610919508813250

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

The idea of rank interaction in factorial experiments when the response is only measured on an ordinal scale was introduced by de Kroon and van der Laan (1981). In this paper, the exact null distributions of test statistics for main effects and rank interaction are compared with their independent asymptotic x2distributions, in particular for experiments with an ordered categorical response using an arbitrary score. Expressions are derived for the exact null variances and covariance of the two statistics for balanced experiments. Simulation is used to make further comparisons between the exact and the asymptotic distributions, and to consider experiments which are not balanced. Some recommendations are made as to when the asymptotic distributions may safely be used

ISSN:0361-0918    

DOI:10.1080/03610919508813251

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor

摘要:

A parallel algorithm for Kalman filtering with contaminated observations is developed. This algorithm is suitable for the parallel computer implementation allowing to treat dynamic linear systems with large number of state variables in a robust recursive way. The implementation is based on the square root version of the Kalman filter. It represents a great improvement over serial implementations reducing drastically computational costs for each state update and avoiding numerical instability problems.

ISSN:0361-0918    

DOI:10.1080/03610919508813252

出版商:Marcel Dekker, Inc.    

年代:1995

数据来源: Taylor