Replacing missing values from a numeric variable with -1

theWaytheWay Member Posts: 4 Contributor I
I don't understand the purpose behind that. What if I'm prepping the data to do a logistic regression? How is the -1 going to affect things?

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist
    Hi,

    I am not sure for a logistic regression. But there are sometimes cases where you e.g. a algorithm did not work and did not give you any result. In this case people sometimes replace with "magic" values. e.g. -1 or -42. Thats one of the common things to do.
    Be careful using it. For some algorithms this -1 makes things wired. E.g. k-NN.

    ~Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.