Avoiding Dummy variable trap

Should we drop any of the dummy variables, just chose one randomly or is there a way to estimate which one is the best to drop?
I’ve assumed the one that has the higher correlation to others but in the lesson we’re dropping random dummies. For example Teenager age category isn’t really correlated to any other category, but Young Adult correlates with Adult, yet we’re chosing Teenager to drop.

Hi there! :blush:
I had the same question and this post helped me to understand the concept. I hope it will help you too!

1 Like