Orange Forum • View topic - lift/uplift/netlift modeling in orange

lift/uplift/netlift modeling in orange

A place to ask questions about methods in Orange and how they are used and other general support.

lift/uplift/netlift modeling in orange

Postby jcress410 » Mon Jan 23, 2012 21:13

For tree based modeling most packages use some data element as the dependent variable. In uplift modeling the goal is to maximize the difference between treatment and control groups.

EDIT (revise and amend)

One way to implement this is with a CART which maximizes the t-statistic for an interaction term between the split (1 = left, 0 = right) and the treatment variable (1 = treated, 0 = control) b_3 below,

y = b_1 * T + b_2 * P + b_3 * T * P

as I understand it, the interaction term (T*P) captures difference in treatment effects between the two sides of a split.

Being new to orange, I'm not sure how to set up a CART to maximize on a given test statistic.

Return to Questions & Support