摘要:

The purpose is to develop interval estimates for differences between treatment means which, when used for testing, would be equivalent to the Waller-Duncank-ratio rule for multiple comparisons. First the intervals are derived from a family of extendedk-ratio Bayes testing rules (exchangeable priors and additive linear losses). Then the intervals are shown to be Bayes also (same priors and squared-error losses). Particularly striking is the dependence of width and location on the between-treatmentsF-ratio rather than the number of treatments. AsFdecreases the newk-ratiot-intervals shrink in width and shift toward zero.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480309

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

A procedure, using the indifference zone formulation, for making simultaneous pairwise inferences aboutkmeans is presented. This procedure has the advantage of offering protection against both Type I and Type II errors. To determine a sample size large enough to afford this protection, least favorable configurations must be determined. This is done in the two- and three-population cases. Tables for the practical application of the procedure to two, three or four means are provided and a method for extending the tables to higher numbers of means is given.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480310

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Modified ratio estimators for the mean are proposed for the case in which the first- and second-phase samples are drawn without replacement and independent of each other. Conditions for these estimators to be more efficient than the classical estimator are investigated. Two methods for evaluating the expectation of the reciprocal of the number of distinct units are presented.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480311

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

A modification of the small sample up-and-down method is studied for the normal, logistic, and extreme value distributions. Two specimens are tested “in series” (i.e., simultaneously) at each stimulus level, and the decision to increase or decrease the stimulus level for the next test is made as soon as thefirstspecimen responds. This procedure can markedly reduce the total test time in certain situations. Overall, the efficiency of the new procedure, as judged using Dixon'sEfcriterion, is sufficiently high to warrant its consideration when overall test time is much more critical than specimen cost.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480312

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480313

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Monte Carlo methods are employed to compare the effectiveness of pair-matched and independent stratified samples for estimating relative risk in the presence of bias. Three approximations to the maximum likelihood estimator for stratified samples suggested by Woolf, Mantel and Haenszel and Birch, respectively [25, 17, 2], are also compared. The sampling model is modified to approximate the practical choices for a researcher and to allow for the loss of unmatchable sample units. The mean square error is always largest for the matched-pairs estimator, while of the stratified estimators, Woolf's consistently produces the smallest MSE, equaled only by Birch's when the samples are equal.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480314

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

The likelihood ratio method for confidence interval estimation of survival or life-time probabilities is investigated for censored data. This approach depends on a constrained product-limit estimator for the survival function. The likelihood method is compared with two alternative methods based on normal approximations to the product-limit estimator for several cases of uncensored and randomly censored samples. The methods are generalized for the ratio of two survival probabilities.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480315

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Four methods for estimating heteroscedastic variances are discussed in this article: the MINQUE introduced by C.R. Rao [7], the AUE introduced by S.D. Horn, R.A. Horn, and D.B. Duncan [4], the average of the squared residuals, and the sample variance. Properties of these estimators, including translation invariance, existence, bias, consistency, existence of negative estimates, and mean square error (MSE) are compared. In particular, it is shown that the AUE has smaller MSE than either the MINQUE or the sample variance in a wide range of situations.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480316

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

For three symmetric distributions and six sample sizes, this article presents estimates of the small-sample variance of Pitman's location-invariant and location-scale-invariant estimators of location. It then compares these two estimators and investigates the closeness of the Cramér-Rao bound when estimators are required to be invariant. Expressions of the formc/(n−d) prove quite effective in fitting variances of Pitman estimators, anddcan be interpreted as the amount of Fisher information “lost.” In terms of relative efficiency, maximum-likelihood estimators and linear combinations of order statistics offer computationally attractive alternatives to the Pitman estimators.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480317

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor

摘要:

Exact one-sided simultaneous confidence bounds are obtained for general regression surfaces with intercepts and functionally independent regressors. For all values of the regressors, the distance from the regression surface to the one-sided bound is a fixed fraction (less than one) of the corresponding distance to the well-known Working-Hotelling-Scheffé two-sided bound. Tables are provided to facilitate the use of such one-sided bounds.

ISSN:0162-1459    

DOI:10.1080/01621459.1975.10480318

出版商:Taylor & Francis Group    

年代:1975

数据来源: Taylor