This is a tricky one, but I think Option B is the way to go. The lift chart is all about improving the model's performance, so it's about the number of events, not accuracy.
I'm leaning towards Option D. Selecting the observations with a 10% response probability should result in 3.14 times greater accuracy compared to a random draw.
I agree with you, Option D seems to be the most accurate. Selecting the observations with a response probability of at least 10% should result in 3.14 times greater accuracy than a random draw of 10%.
I'm not sure about Option C. Selecting the top 10% of the population scored by the model should result in 3.14 times greater accuracy than a random draw of 10%.
I disagree, I believe Option B is the right choice. Selecting the observations with a response probability of at least 10% should result in 3.14 times more events than a random draw of 10%.
Option B seems correct. The lift chart is showing the lift at a certain depth, which means the selected observations should have 3.14 times more events than a random draw of the same size.
Olive
2 months agoCristy
14 days agoDannette
1 months agoElvera
1 months agoMarlon
2 months agoKrystal
8 days agoEmile
9 days agoMozell
11 days agoBrigette
15 days agoErasmo
1 months agoLatrice
2 months agoCristy
24 days agoJerry
27 days agoErasmo
2 months agoLoreta
2 months agoLenna
3 months ago