turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Community Home
- :
- Product Help
- :
- RapidMiner Studio Forum
- :
- New Logistic Regression Operator in Version 7.3

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Get Direct Link
- Email to a Friend
- Report Inappropriate Content

02-10-2017 03:56 PM - edited 02-10-2017 03:57 PM

02-10-2017 03:56 PM - edited 02-10-2017 03:57 PM

I'm glad finally there is an operator in Rapidminer to estimate *standard* logistic regression. My question has to do with the class defined as "success". Let's say I have a binomial label Y with two possible outcome "A" or "B". Which equation is being estimated by RM:

P( Y = "A" | regressors ) = Exp( beta*X) / (1 + Exp(beta*X) )

or

P( Y = "B" | regressors ) = Exp( beta*X) / (1 + Exp(beta*X) ).

Before this operator became available I used the Weka operator W-Logistic. The output of this operator indicated the Class used as "success" (for instance in the output below the modeled class is "false".