摘要:

The level and power of the usual F-test for the r-way layout in analysis of variance (ANOVA) are vitiated when the assumption of equality of error variances is violated. We present procedures, with tables and approximations needed for implementation, which give exact tests with power and level completley independent of the unknown variances. Thus. the assumption of equal variances may be dropped.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489696

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

The use of moving sums (MOSUMS) of recursively calculated residuals (Hack1 [9]) is demonstrated for testing the constancy of the expected value of a series of T independent observations normally distributed with equal variances. The cases of known or unknown expected value are treated separately; in the situation of unknown variance the recursive estimation of the variance is suggested in the course of the sequential application of the test procedure. Conservative probability limits are constructed and an assessment of the effective error probabilities and the power of the method is given in simulation studies. Finally the use of the procedure is demonstrated in an example.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489697

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

This paper examines the problem of determining sample sizes for comparing sample proportions from n independent binomial populations under the null hypothesis that the population proportions are different. For specified precisions and sampling costs, it is shown that the problem of determining optimal sample sizes can be formulated as a convex programming problem. In addition, the problem of obtaining the best precision within a budgeted cost of sampling is examined. This methodology is illustrated with an example application.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489698

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

Letx1,x2, …,Xn, be independent normal variates with a common variance σ2and suppose thatk1of thexiare outliers. That is,k1of thexicome from a population with mean μ + λ while the remainingn−k1of thexicome from a population with mean μ.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489699

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

Accelerated life testing for distributions with hazard rate functions of the formr(t) =Ag(t)+Bh(t) are considered. LetV1, …,Vkbe stress levels larger thanV0—the stress level under normal conditions [V0> 0]—and leta(v) be a nondecreasing function on (0, ∞). We discuss a generalization of the common accelerated models (the power rule model and the Arrhenius model) by assuming that the hazard rate under the stress levelV, is of the form (a(Vt))P(Ag(t) +Bh(t)). The maximum likelihood estimators ofA,BandPfor complete and censored samples are studied. The estimation procedure reduces to a solution of one equation with one unknown parameter. The estimation procedure under the assumption of aging is also described. The asymptotic variance-covariance matrix is given.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489700

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

Transformations are devised to accelerate convergence to normality for maximum likelihood estimators of the reliability of complex systems. Improved confidence intervals follow from probability statements concerning the transformed estimator. Comparisons are made with intervals given by exact procedures and also by generating test results from a system of given reliability, enabling the confidence property of intervals obtained using the transformation to be assessed. Results using other methods, including the log odds transformation for binomial subsystem test data, are also given. The generality of the approach is illustrated by using the transformations to improve confidence intervals in cases where other methods are not appropriate.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489701

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

The quartic exponential (QE) distribution defined by the probability density function of the type

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489702

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

Methods for constructing uniformly most powerful unbiased two-sided tests of the parameters of the gamma distribution and the associated unbiased confidence intervals are discussed. These methods make use of existing results on construction of unbiased confidence intervals for the variance of a normal distribution. An approximation is derived for the conditional distribution of the statisticWn= x/x givenGn= x/β. Tabulations which are convenient in calculating confidence limits for the shape parameter are given.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489703

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

In this note a decision rule presented by Suich and Derringer [12] for aiding model validation is examined and contrasted to others in the literature. These decision rules which lead to preliminary test estimators are compared under squared error loss to the risk of the conventional estimator. Finally an estimator is identified which is uniformly superior to (dominates) the pretest estimator over the entire range of the parameter space.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489704

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor

摘要:

Schafer and Rutemiller [4] presented a solution to the complete ranking problem based on chi-squared random variables. Their solution is unnecessarily restrictive since a general distribution for the ranking variables may be assumed as in Bishop and Dudewicl [2]. and in fact in some cases the exact form of this general distribution may be unknown.

ISSN:0040-1706    

DOI:10.1080/00401706.1978.10489705

出版商:Taylor & Francis Group    

年代:1978

数据来源: Taylor