|
Abstract:
|
A mixture experiment involves blending two or more components in various proportions and measuring one or more response variables on each blend. The slack variable (SV) approach is one possible approach for designing and modeling the data from mixture experiments. With the SV approach, one component is designated the SV, the experiment is designed, and the data are analyzed in terms of the remaining components. A SV proportion is obtained by subtracting from one the sum of the proportions of the remaining components, thus "taking up the slack." With the mixture approach, the experiment is designed and the data are analyzed using all of the mixture components. The presentation considers the situations in which the SV approach has been proposed, and argues in each situation that it is generally preferable to use an appropriate mixture approach.
|
