摘要:

It frequently happens that data to be treated by regression methods or analysis of variance are subject to marked heterogeneity in the error variance. Weighting of the observations, in the least squares sense, is then appropriate but requires that values of the relative weighting factors be available. A method is presented for deriving weight estimates from the residuals of an analysis carried out with any preliminary set of weights. The method is iterative and may therefore be applied with an initial set of unit weights for all observations. Specifically, the method is concerned with column weighting in two-way tables, treated in accordance with an additive model or a more general linear regression model. The effectiveness of the method has been tested by Monte-Carlo techniques. An application of the method to an interlaboratory study of a test method is discussed.

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490144

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490145

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

Through the use of a “predictor-corrector” equation an experimenter may quickly determine the least squares estimates of all the coefficients in a polynomial model after the conclusion of each run, given that an initial set of orthogonal estimates of the coefficients is available. The analysis of variance may similarly be constructed at the conclusion of each added run. The predictor-corrector equation is subject to mild restrictions which are fully met in the usual application of the 2kand 2k–pfactorial designs. Adaptation of the predictor-corrector equation to the case of complimentary blocks of two runs each is described. Examples are given.

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490146

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

摘要:

Given two production processes, the units from which fail in accordance with the Weibull distribution, the problem of selecting the particular process with the larger mean life is considered. Three techniques are considered and three procedures are constructed to show their advantages. An alternative sequential procedure has also been proposed and shown to be valuable in some situations. This paper may be regarded as a generalization of Sobel's earlier work [21] on the exponential distribution.

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490147

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490148

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490149

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490150

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490151

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490152

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1964.10490153

出版商:Taylor & Francis Group    

年代:1964

数据来源: Taylor