0
$\begingroup$

I'm working on a classification problem. I'm trying to build a model which can predict if a bank client will get a loan or not. Some of clients have co-borrower and the majority don't.

I also have information on co-borrowers like salary, etc. but as the majority of clients don't have co-borrower, I have missing values. How can I impute this missing data ?

thank you

$\endgroup$
  • $\begingroup$ Depends on your whole data. But one simple way, just drop co-borrowwer $\endgroup$ – bakka Oct 8 '18 at 9:43
  • $\begingroup$ thank you sir for your response unfortunately, i can't drop co-borrowers because in the case when they exist , client get the loan so I think it's a pertinent information. $\endgroup$ – Haythem Mzoughi Oct 8 '18 at 9:46
  • $\begingroup$ How about just a binary has_coborrower indicator? $\endgroup$ – shadowtalker Oct 8 '18 at 15:25
0
$\begingroup$

As @shadowtalker indicated, a binary features indicating the existence of co-borrower should be helpful anyway. As you certainly know, "imputing missing values" as a procedure for guessing value of missing values is absolutely meaningless as it produces values for something which do not exist. If it's about salary and similar numeric data, just simply put 0 (or the neutral value. Whatever it is.) and if it's about nominal variables, add a new value equivalent to None. at the end, all single customers must have exactly same values for these features.

Also one practical idea could be dividing your data to single borrowers and 2-borrowers at the very beginning of classification so the whole model will become like this: enter image description here

Above model is somehow equivalent to the feature-selection property of some classifiers such that Decision Trees. In a Bayesian Network, such a feature could be the root node.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.