Best-subset analysis with size-correction: SPSS command file generator
This is an outdated version. For the newest version, that also allows conventional best subset analysis (no size correction), click here.
This webpage generates a SPSS 12-13 command file for the best-subset Canonical Variates Analysis/Discriminant Function Analysis with Darroch and Mosimann (1985) size correction.
The output will be saved as file "results.txt" in the working directory and can be analyzed further in any spreadsheet application. This is not an optimal solution and it will work efficiently for a limited number of variables. You can modify this script for your own needs.
Assumptions: your independent variables must conform the SPSS default for naming variables and the sequence must start from 1. For example, if you have 11 variables, the first one must be called var00001 and the last one var00011.
Your datamatrix should not have any variables called v1, v2 ....
Notice: when the command file is generated copy it to the clipboard rather than save it as a text file. Otherwise, extra line breaks will interfere with the SPSS syntax.
The script can be executed as a standalone version once this webpage is downloaded on your computer.
If you requested a substantial number of variables wait until the program finishes calculations and overwrites the content of this page