D- and A- (T-) Optimal Algorithmic Design Creation

The D- and A-optimal design procedures provide various options to select from a list of valid (candidate) points (i.e., combinations of factor settings) that will extract the maximum amount of information from the experimental region, given the respective model that you expect to fit to the data.

Element Name Description
General
Detail of computed results reported Specifies the level of computed results reported. At the Minimal Level of detail, STATISTICA reports various summary statistics for the final design, the final design, and the correlation matrix for the design vectors; if All results is selected, the results also include 2D and 3D scatterplots, summarizing the final design.
Points in final design Enter the number of points for the final design.
Ordering of runs Specifies ordering method for runs in the final design.
Random number seed Optional If random ordering is selected, specify a seed from 1 to 99999999 to create the random sequence of runs for the design.
Number of replicates Add to design option: Number of replicates to add to the design (minimum 0, maximum 300).
Number of blank columns Add to design option: Number of blank columns to add to the design spreadsheet (minimum 0, maximum 300).
Model and Design Specifications
Type of design Select a response surface model with intercept or mixture model without intercept.
Response surface model This option is only applicable if the current design type is a response surface (central composite) design. Specify the type of response surface model to construct when evaluating the optimization criterion; specify the interactions and polynomial effects to include in the model (e.g., quadratic main effects, interactions, etc.).
Mixture model This option is only applicable if the current design type is a mixture response surface design. Specify the model to construct for the mixture design when evaluating the optimization criterion.
Tolerance, mixture total This value is used to verify that the component values for the points in the candidate list sum to a constant ± tolerance (the mixture total, plus or minus the tolerance value). This value is ignored if a response surface design model is selected.
Optimal Design Options
Optimization criterion Specifies whether to use the D or A (T) criterion for selecting the optimal design.

 Select the D-optimal (maximize determinant) option to maximize the determinant of the cross-product matrix X'X, where X stands for the design matrix. Because the determinant of the X'X matrix can be updated quickly, the computations are usually noticeably faster.

 Select the A- (or T-) optimal (minimize trace of X'X inv.) option to minimize the trace of the inverse of X'X, where X stands for the design matrix. If you select this option, the computations may require more time.
Optimization method Specifies the optimization or design search method. Refer to the Electronic Manual for details regarding these techniques and their advantages and disadvantages.
Max number of iterations Specifies the maximum number of iterations that are to be performed.
Initial design Specifies the method for selecting the points for the initial design.
Alpha for singularity Specifies an alpha value that is to be used for computing the X'X matrix for the initial design. Refer to the Electronic Manual for details.
Tolerance, DA impr; 1E- Specifies the negative exponent for a base-10 constant Delta (Delta = 10^-Delta); this value determines when the respective iterative search will terminate, i.e., it is used to determine whether successive iterations have yielded an improvement in the selected optimality criterion.