摘要:

In mixture experiments, the proportions of the components are often constrained by upper and lower bounds. The resulting experimental region is usually an irregular polyhedron. The experimental region, however, can be obtained from a regular simplex by truncation. This technique allows derivation of formulas for the number of boundaries, and the volume, of the experimental region. The formulas for the number of boundaries provide guidance for designing mixture experiments. The formula for volume may be used to determine if small changes in the constraints have a small effect on the experimental region. The volume formula is also used, along with the truncation procedure, to calculate center-of-mass centroids. A new algorithm is given for generation of the vertices of mixture regions constrained by upper and lower bounds.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488110

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The properties of several jackknife-based estimators are investigated in the context of the nonlinear regression model. These estimators are suggested to overcome lack of balance of the design and the nonlinearity of the model. It is shown that a reweighted estimator combined with a modified variance estimate provides a technique that is reasonably robust with respect to outliers, leverage points, and curvature effects. Several examples are presented to illustrate these properties.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488111

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The use of Stein estimation in multiple linear regression is considered. Tables and graphs are presented that compare the prediction mean squared errors of positive-part James-Stein, preliminary-test, reduced, and full-model least squares estimates. The appropriateness of using Stein contraction on possibly extraneous variables is emphasized, and a procedure is presented for evaluating the likely savings in using Stein estimation on the problem at hand. An example is given.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488112

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

A distribution developed previously (Cook and Johnson 1981) has been generalizedu sing Morgenstern's distribution to include more useful dependence structures. The new distribution provides better fits to a uranium survey data set. The effectiveness of Atkinson's (1982) graphical aid for discriminating separate models is also demonstrated in comparing the usual normal model with the new distribution having normal marginal distributions.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488113

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The concentration and entropy measures for categorical data tend to be highly dependent on the choice of response categories. If the categorization of the response variable is arbitiary but it is reasonable to assume an underlying continuous distribution, then an adaptation of the regressionR2measure can be useful for describing multiple association.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488114

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

The binomial failure rate model can be used to estimate the rate of simultaneous failure of more than one component of a system. This review article defines the model, shows how quantities of interest are estimated, and presents checks for lack of fit. Emphasis is on the statistical and computational ideas rather than on the details of the method.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488115

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

This article examines the effects of the choice of design points on calibration in the presence of a simple linear regression model. The distribution of the classical estimator is examined via modes, median, and moments, and the asymptotic variance of this estimator is minimized. The influence of the designs on confidence regions, constructed via Fieller's theorem, is investigated through accuracy curves. A numerical example is presented, and some additional problems are suggested.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488116

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

This article shows that the likelihood ratio test for the equality of location parameters ofk(k≥ 2) two-parameter exponential distributions with unequal scale parameters, in general, depends on the unknown scale parameters under the null hypothesis. A modification of the likelihood ratio test is suggested, which is applicable to Type II censored or complete data. The suggested test is consistent and asymptotically optimal in the sense of Bahadur efficiency. Fork= 2, the power function of the test is obtained, and the test is shown to be unbiased. A set of data is analyzed.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488117

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

摘要:

Consider the following experiment: At stress levels x1< … < xk, n1, …, nkitems are put on test. The failure times of those items failing before timest1*,tk* are recorded, and the remaining failure times are censored. LetP(t,x) denote the probability that an item stressed at levelxfails on or before timet. We consider models of the formP(t,x) = Φ[g(t,x)], where g is a real-valued function specified up to an unknown parameterc, and Φ is an unspecified continuous distribution function. Under these assumptions a Kolmogorov-Smirnov-type test is discussed as a method for estimating c as well asP(to, xo) for specifiedtoand xo. The method is exact and does not require failures at unaccelerated stress levels. Ifn1= … =nk, the method extends to experiments that at each stress level are terminated after a specified number of failures is observed.

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488118

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor

ISSN:0040-1706    

DOI:10.1080/00401706.1986.10488119

出版商:Taylor & Francis Group    

年代:1986

数据来源: Taylor