• Home
  • Software
  • Industries
  • Application Areas
  • Academic
  • Training
  • Support
  • User Community
Free Trial Buy JMP International About Us Contact News and Events Blog Resource Center
  • JMP FAQs
  • System Requirements
  • Alphabetical List of Features
  • Downloads
  • Knowledge Base
  • Books and JMP
  • Report a Problem
  • JMP Administrators
  • Find Your Serial Number or Site ID
  • Why Register JMP?
  • Register JMP

*On SAS Download site

Statistical FAQs

Can I get discriminant function scores or function coefficients?

Version 5.1.2 and higher: Yes. Discriminant analysis is available with the Analyze > Multivariate Methods > Discriminant menu selection. To get the discriminant scores, select Score Options > Save Formulas.

Version 4 and Version 5.0: Yes. See pp. 334-335 of JMP Statistics and Graphics Guide, V4 or pp. 333-334 of JMP Statistics and Graphics Guide, V5 or pp. 513-514 of JMP Statistics and Graphics Guide, V6 for a general description and detailed formulas. If you select Save Discrim from the Manova Fit pull-down menu, you'll get additional columns in the data table with the Mahalanobis distances and posterior probabilities. The values in the DIST variables are the Mahalanobis distances of each observation to the group centroids and could be called the scores on the discriminant functions. Each of these columns has a formula associated with it. If you look at the formula for the DIST variables, you see the coefficients of the functions used in JMP (which are not the same as those in SAS). Details of the computations can be found in Appendix A of the JMP Statistics and Graphics Guide.

Version 3: Yes. See pp. 252-253 of the JMP Statistics and Graphics Guide, V3 or pp. 255-256 of JMP Statistics and Graphics Guide, V3.1 for a general description and detailed formulas. If you select Save Discriminant Formulas from the $ menu, you'll get additional columns in the data table with the Mahalanobis distances and posterior probabilities. The values in the DIST variables are the Mahalanobis distances of each observation to the group centroids and could be called the scores on the discriminant functions. Each of these columns has a formula associated with it. If you look at the formula for the DIST variables, you see the coefficients of the functions used in JMP (which are not the same as those in SAS). Details of the computations can be found in Appendix A of the JMP Statistics and Graphics Guide.

 


FAQ # 465
Last Updated: 2007 Sep 05

Browse JMP FAQs

General FAQs

Statistical FAQs


Support Documents

JMP FAQs

Knowledge Base

System Requirements

Alphabetical List of Features

Known Issues


Contact JMP Support

Report a Problem

Update a Report

support@jmp.com

919.677.8008 (US)

International Support via Worldwide SAS Offices

 
SAS | JMP is a business unit of SAS. SAS Campus Drive, Building S, Cary, NC, 27513 • Phone: 1.919.677.8000
© 2008 SAS Institute, Cary, NC, USA. All Rights Reserved. Terms of Use • Privacy Policy