scientific article; zbMATH DE number 1107644
The response surface for the process mean is quite straight-forward, and its properties are well defined. The response surface for the process variance is generated by using the variance operator and requires that the variance-covariance structure of the noise factors as they behave in the process be either known or well estimated. This paper develops the prediction properties of the process variance response surface under two scenarios: when the single linear model is correctly specified, and when it is underspecified. Section 2 outlines the bias and the variance for the estimate of the process variance when the model is correctly specified. This section concludes with the development of an integrated mean squared error criterion which allows an analyst to compare the ability of several designs to estimate the process variance. Section 3 extends these results to the situation where the model is underspecified. Section 4 applies the integrated mean squared error criterion to the cake mix example used by \textit{G. E. P. Box} and \textit{S. Jones} [Total Qual. Managem. 3, 265-282 (1992)]. Section 5 gives our conclusions and recommendations for future research. For the entire collection see [Zbl 0867.00026].
- scientific article; zbMATH DE number 1121367
- A multivariate process variability monitoring based on individual observations
- On relations between prediction error covariance of univariate and multivariate processes
- scientific article; zbMATH DE number 1029179
- COMPONENTS OF PREDICTION ERRORS FOR A STATIONARY PROCESS WITH ESTIMATED PARAMETERS
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4373583)
