Best Subset Discriminant Analysis Options in GDA

Select the Best subsets selection of predictor effects as the Model building option on the Quick specs dialog box - Advanced tab or the GDA Models Wizard--Extended Options dialog box - Advanced tab to display the options described here. Remember that the total number of all possible subsets (that need to be reviewed by STATISTICA) can become excessively large when there are many effects in the model and many large subset sizes are being considered (via the Start and Stop fields).

Wilks' lambda

Select the Wilks' lambda option button to use the Wilks' Lambda value as the criterion for choosing the best subset of predictor effects. The Wilks' Lambda statistic for the overall discrimination is computed as the ratio of the determinant of the within-groups variance/covariance matrix over the determinant of the total variance covariance matrix:

Wilks' Lambda = det(W)/ det(T)

The F approximation to Wilks' Lambda is computed following Rao (1951).

Analysis misclass

Select the Analysis misclass. option button to use the misclassification (error) rate value of analysis (training or learning) sample data as the criterion for choosing the best subset of predictor effects; the misclassification error rate is computed as the number of misclassified observations divided by total number of observations.

Crossval. misclass.

Select the Crossval. misclass option button to use the misclassification (error) rate value of cross-validation (or test) sample data as the criterion for choosing the best subset of predictor effects; the misclassification error rate is computed as number of misclassified observations divided by total number of observations.

Start, Stop

Enter values in the Start and Stop fields to determine the sizes of the subsets that will be considered during the search through all possible subsets. STATISTICA will begin the search with the subset size specified in the Start field, and will terminate the search after all subsets of the size specified in the Stop field have been evaluated.

Subsets to display

Enter a value in the Subsets to display field to determine the number of subsets of each size that will be displayed in the Summary of best subset search spreadsheet (available when you click the Summary of best subset search button on the GDA Results - Quick tab); for example, if you specify 10 in this field, then you can later review the 10 best subsets (for each subset size) according to the chosen criterion for each size of subset that was considered (see options Start, Stop above).

See also GDA - Index.