Age is coming out as a really strong predictor compared to other variables. This is a classification problem, the dependent variable is a (0/1)


Hard to know exactly without knowing what kind of model you're using, but it would be worth repeating the model without age to see what you get. It could be that you're missing important correlations by adjusting for age. Collider bias is one potential issue here.

I would report univariable correlations and a multivariable model in this case.

| cite | improve this answer | |
  • $\begingroup$ OP is estimating a Gradient Boosting algorithm it seems so may not be predominantly interested in consistent estimates of some coefficient, but you're right that some context would be helpful $\endgroup$ – Mark Verhagen Apr 29 at 9:17
  • $\begingroup$ I assumed they were just running some multivariable logistic regression to predict the 0/1 dependent variable $\endgroup$ – H. Green Apr 29 at 9:21
  • $\begingroup$ Yes, I'm using gradient boosting algorithm to classify, building a look-alike model. When I look at the feature importance plot the age variable just overshadows all other variables. I'm not sure if it is right using a such a strong predictor, would it over shadow any other variable? I am getting a good AUC of 0.787.. Please let me know if you need any other details. $\endgroup$ – Gaurav Manjunath Apr 29 at 10:01

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.