Multivariate Nominal Logistic Regression- Prediction Expression

[#1]

I am using the Fit Model platform to fit three continuous variables to a binary outcome.  There is an option to "Save Probability Formula" which then outputs onto my original data data, and then populates probabilities and then classifications when I input new rows of data for my three variables.  However, I do not see an option to Show Prediction Expression.

e.g. 

Variables:  Temperature, Length, Weight

Outcome:  Pass/Fail

Is there a function on the Fit Model platform for this type of analysis that is equivalent to the "Show Prediction Expression" normally seen with Standard Least Squares?  I would like to be able to document that.  I can imagine taking the data of my probabilities and regressing it against my variables with using the Fit Model Outcome again, taking the prediction expression from that, and then manually classifying the outcomes, but it's an extra step and probably not the same.  

[#2 · Accepted solution]

The formula is for the probability of each outcome.  The prediction expression would depend on what probability you choose for a cutoff for classification.  So, if you use the default 50% cutoff, then the prediction expression would be Yes if prob >=50% and No if prob<=50%.  In general, I don't think the 50% is appropriate (unless the costs associated with misclassifications are equal for both types of errors), but the prediction of a classification is just that simple formula.  The real formula is the probability formula, and it shows as the formula property for that column.

[#3]

Thank you! I didn't know that the Formula would be embedded and viewable within the column of the data table.