Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Best practices for modeling on an ordinal response target

Reply
Senior User MZM
Senior User
Posts: 1

Best practices for modeling on an ordinal response target

[ Edited ]

Hi Modellers,

 

I am trying to build a model on a dataset with 2 million observations and an ordinal response variable with 53 levels with a normal distribution ranging from -26 to +26.

Both SAS/Base 9.4 and Enterprise Miner 13.2 are available for use. 

 

I am looking for any suggestions on modeling techniques that could be used especially in EM. 

Was also wondering if there's a way to use cumulative logit link function in EM.

Does that make sense to consider the response variable interval and then to use GLM?

 

Thanks,

M.

Trusted Advisor
Posts: 1,605

Re: Best practices for modeling on an ordinal response target

[ Edited ]

Unless your 53 levels are highly non-linear, I would treat them as a continuous variable and perform partial least squares regression modelling (which in my opinion is probably the best way to model 900 independent variables) in PROC PLS. I do not know if this is available in Enterprise Miner as I don't use it. I would not use PROC GLM with 900 independent variables.

SAS Super FREQ
Posts: 269

Re: Best practices for modeling on an ordinal response target

When you use the Regression node in Enterprise Miner with an ordinal target, it does use the cumulative logit link function.

Ask a Question
Discussion stats
  • 2 replies
  • 166 views
  • 2 likes
  • 3 in conversation