Predict new observations for Analyze Mixture Design

Stat > DOE > Mixture > Analyze Mixture Design > Prediction
Use when you want to calculate and store predicted response values for new design points.
New Design Points (columns and/or constants)
Components
Enter the new observation for each component in the same order that it is entered in the model. You can enter one numeric component value for each component, or enter the columns or constants in which they are stored. Columns must have the same number of rows. The number of components must match the number in the design. Component values must be entered in the same units as they are displayed in the worksheet. Components that are displayed in the worksheet as amounts must sum to the common total or the total used for prediction in Mixture Amount below. Components that are displayed in the worksheet as proportions or pseudocomponents must sum to one.
Process Variables
Enter the new observation for each process variable in the same order that each process variable is entered in the model. You can enter one text or numeric process variable value for each process variable, or enter the columns or constants in which they are stored. Each value of a process variable must equal one of the values in the design. Process variable values must be entered in the same units as they are displayed in the worksheet. Columns must have the same number of rows. The number of process variables must match the number in the design. You must enclose text values in double quotes (e.g., "High").
Mixture Amount
Enter the numeric mixture amount, or enter the column in which they are stored. The value must equal one of the mixture amounts in the design.
Blocks
You do not have to enter a block value. You can enter one text or numeric block value for the block, or enter the column in which they are stored. Each block value must equal one of the block values in the design. You must enclose text values in double quotes (e.g., "Monday").
Confidence level

Enter the level of confidence for the confidence intervals and the prediction intervals. Usually, a confidence level of 95% works well. A 95% confidence level indicates that if you performed the experiment 100 times, the confidence interval would contain the mean response approximately 95 of the times. Similarly, the prediction interval indicates that you can be 95% confident that the interval contains the value of a single new observation.

Storage
For more details on any of these statistics, go to Prediction table for Analyze Mixture Design.
Fits
Minitab stores the fitted values for new observations in the column PFit.
SE of fits
Minitab stores the estimated standard errors of the fitted values in the column PSEFit.
Confidence limits
Minitab stores the lower and upper limits of the confidence interval for the prediction in the columns CLimLo and CLimHi.
Prediction limits
Minitab stores the lower and upper limits of the prediction interval in the columns PLimLo and PLimHi.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy