# Is it a good idea to train with a feature which value will be fixed in future predictions?

I am facing a regression problem and I have one feature that has some relevant correlation with the output. The value of this feature will be fixed in ALL the predictions I will use this model for.

Should I keep it in my model or not?

Thanks

• For prediction (future scoring), are you using the same population but just at a different time? – Vishal Apr 24 '16 at 17:32
• No, I will always use new instances. – hipoglucido Apr 25 '16 at 12:43

x_1  x_2  y