|
Analyze Mixture DesignMixture Regression |
For each term in the model, there is a coefficient. Use these coefficients to construct an equation representing the relationship between the response and the design variables (components, process variables, and amount variable).
To use this equation, put in the component proportions and the coded process variable values and calculate the predicted response. Because the coefficients are estimated using component proportions and coded process variables, putting the components in pseudocomponents or amounts or uncoded process variable values into this equation would generate incorrect predictions.
Example Output |
Estimated Regression Coefficients for Flavor (component proportions)
Term Coef SE Coef T P VIF Emmentha 500 141.50 * * 24937 Gruyere -89 54.85 * * 613 Wine 395 143.73 * * 43038 Emmentha*Gruyere 1200 375.92 3.19 0.019 2655 Emmentha*Wine -1676 593.83 -2.82 0.030 102454 Gruyere*Wine -869 409.98 -2.12 0.078 7888 Emmentha*Temperat -572 141.50 -4.04 0.007 24937 Gruyere*Temperat 263 54.85 4.79 0.003 613 Wine*Temperat -568 143.73 -3.95 0.008 43038 Emmentha*Gruyere* Temperat -1440 375.92 -3.83 0.009 2655 Emmentha*Wine*Temperat 2388 593.83 4.02 0.007 102454 Gruyere*Wine*Temperat 1433 409.98 3.50 0.013 7888
* NOTE * Coefficients are calculated for coded process variables. |
Interpretation |
|
For the fondue data, the regression equation is:
Flavor = 500(Emmentha) - 89(Gruyere) + 395(Wine) + 1200(Emmentha*Gruyere) - 1676(Emmentha*Wine) - 869(Gruyere*Wine) - 572(Emmentha*Temperat) + 263(Gruyere*Temperat) - 568(Wine*Temperat) - 1440(Emmentha*Gruyere*Temperat) + 2388(Emmentha*Wine*Temperat) + 1433(Gruyere*Wine*Temperat)