Abstract
The paper describes an experimental procedure to choose the values for a multivariate vector x, under these conditions: average of Y(x) equal to a target value and least variance of Y(x), linked to x by a 2nd order model, with a heteroschedastic error. The procedure consists of two steps. In the first step an experimental design (we consider a three level full factorial design, for simplicity) is performed in the feasible space X of the control factors to estimate the parameters characterizing the response surface of the mean. Then a second experimental design is performed on a target set A, subset of X satisfying the condition on the average of Y(x). This second step determines the choice of x using a classification criterion based on the ordering of the sample mean squared errors. In both steps the model parameters are estimated by an iterative method.
| Original language | English |
|---|---|
| Title of host publication | First joint meeting of the Société Francophone de Classification and the Classification and Data Analysis Group of the Italian Statistical Society, Book of Short Papers |
| Pages | 369-372 |
| Number of pages | 4 |
| Publication status | Published - 2008 |
| Event | First joint meeting of the Société Francophone de Classification and the Classification and Data Analysis Group of the Italian Statistical Society, Book of Short Papers - Caserta Duration: 11 Jun 2008 → 13 Jun 2008 |
Conference
| Conference | First joint meeting of the Société Francophone de Classification and the Classification and Data Analysis Group of the Italian Statistical Society, Book of Short Papers |
|---|---|
| City | Caserta |
| Period | 11/6/08 → 13/6/08 |
Keywords
- Iterative Generalized Least Squares
- Optimal Experimental Conditions