All Possible Subsets for DDA and/or PDA

[Leah Anderson]

Does anyone have an SPSS syntax that does "all possible subsets"
analysis for either Descriptive Discriminant Analysis or Predictive
Discriminant Analysis?

=====================
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

[Dale Glaser]

Leah, I found such when I needed to do all subsets regression for dominance analysis; the only minor tweaking you may need to do is for the directories (e.g., C:\temp), etc.....but otherwise works great..it can be found at the following:


http://www.spsstools.net/Syntax/RegressionRepeatedMeasure/DoAll-SubsetsRegressions.txt

 

Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/Alliant
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

---

From: Leah Anderson <[hidden email]>
To: [hidden email]
Sent: Monday, October 14, 2013 10:18 AM
Subject: All Possible Subsets for DDA and/or PDA

Does anyone have an SPSS syntax that does "all possible subsets"
analysis for either Descriptive Discriminant Analysis or Predictive
Discriminant Analysis?

=====================
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

[David Marso]

I suspect that with a large number of variables and a large number of cases, that code will take a VERY long time to run.
I would modify that to use a CORRELATION matrix for input to the MATRIX language.
I would use MATRIX to generate indices for the combinations and SWEEP the desired columns to get the regression coefficients, R square etc.
Wouldn't be that hard to modify it for other linear models such as DSC etc...

Dale Glaser wrote

Leah, I found such when I needed to do all subsets regression for dominance analysis; the only minor tweaking you may need to do is for the directories (e.g., C:\temp), etc.....but otherwise works great..it can be found at the following:


http://www.spsstools.net/Syntax/RegressionRepeatedMeasure/DoAll-SubsetsRegressions.txt

 
Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/Alliant
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com


________________________________
 From: Leah Anderson <[hidden email]>
To: [hidden email] 
Sent: Monday, October 14, 2013 10:18 AM
Subject: All Possible Subsets for DDA and/or PDA
 

Does anyone have an SPSS syntax that does "all possible subsets"
analysis for either Descriptive Discriminant Analysis or Predictive
Discriminant Analysis?

=====================
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