@javadba wrote:
I was looking at an older post on GBM parameter tuning techniques on Housing dataset: https://www.analyticsvidhya.com/blog/2016/02/complete-guide-parameter-tuning-gradient-boosting-gbm-python/
What I noticed is that a number of numerical fields that would seem useful for prediction purposes were scrapped in favor of flags indicating they were present or not. Here is one of them:
- Loan_Amount_Submitted_Missing created which is 1 if Loan_Amount_Submitted was missing else 0 | Original variable Loan_Amount_Submitted dropped
I would think the Loan amount would be huge for predicting loan conversion rate. So why was that field dropped?
Posts: 1
Participants: 1