Answer Tree 3.1 vs SPSS add on Decision Tree

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Answer Tree 3.1 vs SPSS add on Decision Tree

Stefan Michel-2
I understand that SPSS Decision Tree will be
available in Fall 2008. Are the features similar to Answer Tree 3.1.?


+++++++++++++++++++++++++++++++++++
Dr. Michel | Partner GmbH
Oberer Chämletenweg 42e
CH-6333 Hünenberg See
www.michel-partner.ch
+++++++++++++++++++++++++++++++++++

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Answer Tree 3.1 vs SPSS add on Decision Tree

Guerrero, Rodrigo
I just purchased Decision Tree for SPSS 16 and it is not exactly the same.  My version does not allow for pruning and it does not have as many option available.  I don't know what v.17 is like.




Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Stefan Michel
Sent: Thursday, September 18, 2008 2:53 AM
To: [hidden email]
Subject: Answer Tree 3.1 vs SPSS add on Decision Tree

I understand that SPSS Decision Tree will be
available in Fall 2008. Are the features similar to Answer Tree 3.1.?


+++++++++++++++++++++++++++++++++++
Dr. Michel | Partner GmbH
Oberer Chämletenweg 42e
CH-6333 Hünenberg See
www.michel-partner.ch
+++++++++++++++++++++++++++++++++++

=====================
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
Reply | Threaded
Open this post in threaded view
|

Decision Analysis

mpirritano
In reply to this post by Stefan Michel-2
I'm glad this was brought up. I've been wanting to ask the list about
the general feeling towards decision trees and decision analysis. I've
just been introduced to this area of research and have been looking at
another software package. It seems to me that decision analysis requires
what in traditional statistics would be overly liberal barely tenable
assumptions. The assumptions I'm referring to are parameter values
supposedly culled from meta-analyses or public databases. I mean, with
inferential statistics we're just hoping that assumptions required to
interpret the test hold, and then with decision analysis we're taking
findings from meta-analyses that usually don't explain very much of the
variance (tell me if I'm wrong, but don't most findings from
meta-analyses only explain maybe 25 or 30% of the variance if you're
lucky) and using those figures, ignoring the error associated with them,
to calculate the odds of different outcomes. It seems that at least if
SPSS included such a module it would be possible to base the analysis on
one's data.

I should say that I've just started to look at this area so I could just
be seeing things with the ignorant oversimplified eyes of the novice. It
that's the case I'm sure listers will be willing to point that out.

Thanks
Matt

Matthew Pirritano, Ph.D.

=====================
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
Reply | Threaded
Open this post in threaded view
|

Re: Answer Tree 3.1 vs SPSS add on Decision Tree

Oliver, Richard
In reply to this post by Guerrero, Rodrigo
Version 17 is essentially exactly the same as version 16.

To clarify some potential confusion, SPSS has had a "tree" add-on option since version 13. The name changed from "Classification Trees" to "Decision Trees" in version 17.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Guerrero, Rodrigo
Sent: Thursday, September 18, 2008 10:02 AM
To: [hidden email]
Subject: Re: Answer Tree 3.1 vs SPSS add on Decision Tree

I just purchased Decision Tree for SPSS 16 and it is not exactly the same.  My version does not allow for pruning and it does not have as many option available.  I don't know what v.17 is like.




Rodrigo A. Guerrero | Director Of Marketing Research and Analysis | The Scooter Store | 830.627.4317


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Stefan Michel
Sent: Thursday, September 18, 2008 2:53 AM
To: [hidden email]
Subject: Answer Tree 3.1 vs SPSS add on Decision Tree

I understand that SPSS Decision Tree will be
available in Fall 2008. Are the features similar to Answer Tree 3.1.?


+++++++++++++++++++++++++++++++++++
Dr. Michel | Partner GmbH
Oberer Chämletenweg 42e
CH-6333 Hünenberg See
www.michel-partner.ch
+++++++++++++++++++++++++++++++++++

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

=====================
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
Reply | Threaded
Open this post in threaded view
|

V13 / V15 compatibility question: Error message

King Douglas
In reply to this post by mpirritano
We use KeySurvey for some online surveying tasks.  They recently change their download parameters.  Now, when an SPSS 13 user successfully downloads a KeySurvey file they can open and work with it, but when trying to save it, the get the following error message:

SPSS 13 for Windows R2 (3820)

The "Save" or "Save as" command failed.  Unsaved recent changes will be lost.  Click "OK" to reopen the current file or Cancel and copy your changes to the Clipboard.

This is in the absence of any changes.  One can open and save the file using SPSS 15, but that doesn't fix the problem for SPSS 13 users.  I haven't experienced any other problems in sharing data files created in SPSS 15 with V13 users.

My hunch is that it has something to do with the V15 capacity for long value labels, although there are none that long in the data file under discussion. 

We don't have the budget to upgrade everyone.  I haven't yet talked to KeySurvey about this.

Any ideas?

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