Lift Chart analysis - Help

Hi all!

I started using Knime a couple of weeks ago and I am doing a logistic regression as exercise to understand, from a customer database, who is going to redeem a specific coupon online.

Now, I did my analysis and I arrived at the lift chart, I was wondering if someone can help me understand it so I can explain it in a correct way to my professor.

This is the lift chart without any math formula added or variable added/removed:

 

And this is the lift chart with the math formula I've added and without the insignificant variables:

How I can effectively explain these charts? Also I noticed that my accuracy from the confusion matric get down when I remove "insignificant" variables and add math formulas. Is this an indicator that my model is wrong and I should add different variables?

Many thanks to everyone can help on this one!

Lara

 

 

Hi Lara,

here is a very nice explaniation of lift charts: https://paultebraak.wordpress.com/2013/10/31/understanding-the-lift-chart/

Best, Iris

Iris, the page you link to really describes the plot type called the “Cumulative Gain Chart” in KNIME. Some people do refer to that as a “lift chart” but that’s different than what KNIME calls a Lift Chart.

The images on LaraZ’s original post don’t show up for me (maybe they are removed after some time?) so I don’t know exactly what the question was. But if they are asking for an explanation of what KNIME calls a Lift Chart, the article you linked to won’t really help them.

Respectfully,
Mark