摘要:

Extrapolation of a fitted regression equation beyong the range of the given data can lead to seriously biased estimates if the assumed relationship does not hold in the region of extrapolation. This is demonstrated by some examples that lead to nonsensical conclusions. Extrapolation from fitted polynomials and from multiple regressions, extrapolation to other environments, and some related topics are also considered. The Inadequacy for extrapolation of the assumed form of the fitted relationship can not always be established by a statistical analysis. Thus, extrapolation can not be supported on statistical grounds alone; It must be justified by physical considerations. Even if the assumed form of the relationship is correct, the extrapolation, though not biased, may be quite imprecise. This can be especially serious when the region of extrapolation is far from the region of the given data, and/or the model is a high-order polynomial or a multiple regression developed from data involving high correlations among the Independent variables.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980791

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

This paper investigates the economic design of X̄-charts with both warning and action limits, based on a widely studied process model. The optimization is accomplished by a three-dimensional pattern search. Various comparisons are made among the minimum cost designs of X̄-charts with and without warning limits and of Cusum charts. No appreciable differences are found. X̄-charts with warning limits are thus recommended for practical application as they are much easier to handle than Cusum charts, and they provide more psychological protection than X̄-charts with action limits only. It is proposed that in application the band enclosed by the warning limits should be 0.85 times as wide as the band enclosed by the action limits. A simplified economic scheme for the design of control parameter values is also suggested.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980792

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

Graphs are presented to provide quick and accurate 5% and 1% tests of the equality of two correlation coefficients based on equal-size samples. For unequal samples, the harmonic mean 2n1n2/(n1+n2) provides a good compromise sample size for using the graphs.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980793

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

In this paper the question of what is the best setting for the target value of an industrial process is explored. A simple graphical procedure is described which takes into account the regular and reduced selling prices, the give-away cost, and the process variability. The specific problem studied is as follows. Individual items are produced continuously from an industrial process. Each item is checked to determine whether it satisfies a critical lower (or upper) specification limit. If it does, it is sold at the regular price; if it does not, it is sold at a reduced price. The target value of a process is generally set somewhat above the lower specification limit (or below the upper specification limit). The further the target is set from the specification limit on the safe side, the lower the proportion of rejected items. However, there is an offsetting cost (sometimes called the give-away cost) which restricts the extent to which the target value can be profitably adjusted in this direction.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980794

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

The principles underlying the recovery of interblock information in the analysis of data from balanced incomplete block experiments are described. A novel method of estimating the ratio of block variance to plot variance, needed in the analysis, is introduced.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980795

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

A computer program is presented that may be used to develop specific sequential variables sampling plans. The program was motivated by the following problem situation. A producer is required to make the following statement about a lot,“With 100γpercent confidence, at least 100P percent of the items in the lot are within specifications.”Although this statement is in terms of a tolerance interval, in effect, it comprises an acceptance sampling plan, since failure to meet the criterion results in rejection of the lot. Since sample sizes are not specified in advance, it may be tempting for the producer to begin with a few items, and then to take additional samples if the criterion is not met. This sequential approach, of course, increases the consumer's risk from an acceptance sampling plan viewpoint. A plan is developed that permits the sequential approach but keeps the consumer's risk at a desired level.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980796

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

摘要:

Mixture experiments are being applied with greater frequency in recent years. A problem confronting the user is data display. In many situations, a response surface can be drawn on the three-component simplex to illustrate the effects of the mixture variables. This paper presents a computer program that uses the printer as a plotter to display response surface contours on a three-component mixture space.

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980797

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980798

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980799

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor

ISSN:0022-4065    

DOI:10.1080/00224065.1977.11980801

出版商:Taylor&Francis    

年代:1977

数据来源: Taylor