Design Criterion for Estimating Mean and Variance Functions

Abstract

In an industrial process, the proper objective is to find the optimal operating conditions with minimum process variability around the target. Vining and Myers (1990) suggest to use the separate model for the mean response and the process varian linear predictor τ_i = log σ_i² is unknown and should be estimated. Noting that the variance of τ_i is heterogeneous, another appropriate D‐opportunity criterion D₃ based on the method of generalized least squares is proposed in this paper.

Citation

Lim, Y.B. (2000), "Design Criterion for Estimating Mean and Variance Functions", Asian Journal on Quality, Vol. 1 No. 1, pp. 32-37. https://doi.org/10.1108/15982688200000005