I'm trying to perform an analysis using IBM SPSS Decision Trees and wondering about some sort of restriction on the number of independent variables, since the data I use is large and may contain more than 1,000 independent variables.
The product's documentation did not list an upper limit, so if anyone knows, I would appreciate it if you could let me know.

Thank you.

------------------------------
清水 瞭
------------------------------

#SPSSStatistics

Hi. I do not know of limitations for the TREE procedure other than these:

• CHAID and Exhaustive CHAID: A categorical dependent variable may not have more than 126 categories. If the dependent variable is categorical, then the limit for a categorical predictor is also 126 categories.
• CRT: A nominal predictor may not have more than 32 categories.
• QUEST: If a predictor is nominal, then the limit for the dependent variable (which must be nominal) is 127 categories. A nominal predictor may not have more than 25 categories.

Did you get an error message about a limitation? If not, then I would assume that the model is limited only by available memory.

------------------------------
Rick Marcantonio
Quality Assurance
IBM
------------------------------

Original Message:
Sent: Mon May 23, 2022 11:10 PM
From: R.. Sa
Subject: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

I'm trying to perform an analysis using IBM SPSS Decision Trees and wondering about some sort of restriction on the number of independent variables, since the data I use is large and may contain more than 1,000 independent variables.
The product's documentation did not list an upper limit, so if anyone knows, I would appreciate it if you could let me know.

Thank you.

------------------------------
清水 瞭
------------------------------

#SPSSStatistics

Original Message:
Sent: 5/24/2022 12:07:00 PM
From: Rick Marcantonio
Subject: RE: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

Hi. I do not know of limitations for the TREE procedure other than these:

• CHAID and Exhaustive CHAID: A categorical dependent variable may not have more than 126 categories. If the dependent variable is categorical, then the limit for a categorical predictor is also 126 categories.
• CRT: A nominal predictor may not have more than 32 categories.
• QUEST: If a predictor is nominal, then the limit for the dependent variable (which must be nominal) is 127 categories. A nominal predictor may not have more than 25 categories.

Did you get an error message about a limitation? If not, then I would assume that the model is limited only by available memory.

------------------------------
Rick Marcantonio
Quality Assurance
IBM
------------------------------

Original Message:
Sent: Mon May 23, 2022 11:10 PM
From: R.. Sa
Subject: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

I'm trying to perform an analysis using IBM SPSS Decision Trees and wondering about some sort of restriction on the number of independent variables, since the data I use is large and may contain more than 1,000 independent variables.
The product's documentation did not list an upper limit, so if anyone knows, I would appreciate it if you could let me know.

Thank you.

------------------------------
清水 瞭
------------------------------

#SPSSStatistics

Hi, Rick & Jon.

Sorry for the late reply and thank you for your answers and advice!
Your answers have cleared up my questions for so far.
If I get any error messages, etc., I will consult you again.

Thanks a lot!!


------------------------------
R.. Sa
------------------------------

Original Message:
Sent: Tue May 24, 2022 12:36 PM
From: Jon Peck
Subject: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

AFAIK, agreeing with Rick, there is no hard limit on the number of variables you can use with TREES, although the runtime would be substantial.  But I would be cautious if you are aiming at selecting just a small number out of 1000 or so.  Trees are notoriously unstable, so be sure to do aggressive cross validation.

--

Jon K Peck
jkpeck@gmail.com

Original Message:
Sent: 5/24/2022 12:07:00 PM
From: Rick Marcantonio
Subject: RE: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

Hi. I do not know of limitations for the TREE procedure other than these:

• CHAID and Exhaustive CHAID: A categorical dependent variable may not have more than 126 categories. If the dependent variable is categorical, then the limit for a categorical predictor is also 126 categories.
• CRT: A nominal predictor may not have more than 32 categories.
• QUEST: If a predictor is nominal, then the limit for the dependent variable (which must be nominal) is 127 categories. A nominal predictor may not have more than 25 categories.

Did you get an error message about a limitation? If not, then I would assume that the model is limited only by available memory.

------------------------------
Rick Marcantonio
Quality Assurance
IBM

Original Message:
Sent: Mon May 23, 2022 11:10 PM
From: R.. Sa
Subject: Hi, Does anyone know how many independent variables can IBM SPSS Decision Trees contain?

I'm trying to perform an analysis using IBM SPSS Decision Trees and wondering about some sort of restriction on the number of independent variables, since the data I use is large and may contain more than 1,000 independent variables.
The product's documentation did not list an upper limit, so if anyone knows, I would appreciate it if you could let me know.

Thank you.

------------------------------
清水 瞭
------------------------------

#SPSSStatistics