The classification algorithm SPSS uses is given in its Algorithms documentation. In recent versions, that can be reached off the Help menu.
In the simplest case (equal priors, pooled covariance matrix), a case is classified into the group whose centroid is nearest. In the space of the canonical discriminant functions, it's a simple Euclidean distance.
Jonathan Fry
SPSS Inc.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of pschatz
Sent: Wednesday, March 26, 2008 9:12 AM
To:
[hidden email]
Subject: Group cut-offs in Discriminant Analysis
I understand that group centroids are the mean of the discriminant groups,
but I am not sure how one calculates actual cut-offs for the groups.
I found one reference, Klecka (1908) stating: for the one-Function case, the
cut-off point between two groups is one half the sum of the two group
centroids, but do not have the original (a Sage book) and was wondering
how/if anyone had experience with calculating cut-offs.
Thanks,
Philip Schatz
Saint Joseph's University
--
View this message in context:
http://www.nabble.com/Group-cut-offs-in-Discriminant-Analysis-tp16301495p16301495.htmlSent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
