SAS Enterprise Miner - Regression Node - missing decision prior

[woiya]

Hi,

pretty new to the modelling world but I seem to get intermittent errors whenever I run regression node (logistic regression) in SAS Enterprise Miner.

The path is easy enough: Data sources - Data Partition - Impute - Transform Variables - Regression

The error is specifically around missing Dec Prior, but I don't want to set any decision prior as the data is not under/oversampled.

However, I just added a new variable in the dataset and re-run the exact same process, it didn't come up with an error this time.

Anyone knows what I'm missing or not seeing here?

Any insight will be appreciated 🙂

Here's the log:

NOTE: SCL source line. 11516 TARGET_CHURN="1"; COUNT=12534; DATAPRIOR=0.18855582551658; TRAINPRIOR=0.18855582551658; DECPRIOR=; DECISION1=1; DECISION2=0; - 22 ERROR 22-322: Syntax error, expecting one of the following: a name, a quoted string, a numeric constant, a datetime constant, a missing value, INPUT, PUT.

[MikeStockstill]

Hello-

The Regression node log is not showing a very helpful message, unfortunately.  

Sometimes you can force the flow to re-execute in order to clear the error.  Follow these steps:

  - Go to the first node in the flow, and change the Rerun property from No to Yes.

  - Run the flow.  With Rerun set to Yes, running the flow causes it to re-execute.

  - Optionally, change the Rerun property back to no.

If you still get an error, then turn on the MPRINT option, re-execute the flow to create a model package, and send the model package to technical support.  Here are some links with details.

58396 - Specify the MPRINT option in SAS® Enterprise Miner(tm) Project Start Code to display additio...

46764 - Creating a model package (SPK file) in SAS® Enterprise Miner(tm)

41211 - Techniques for testing, solving, or reporting problems that occur when using SAS® Enterprise...

Have a good week!