2 separate models (smoker & non-smoker) vs 1 model for both

Hi All,

I have tried two experiments:
A) 2 separate linear regression models (smoker & non-smoker)
B) 1 single linear regression model for both smoker & non-smoker

It was not clear based on each model’s respective loss functions to determine whether two separate linear regression models, one for smokers and one of non-smokers, better than a single linear regression model.

Since that is one of the posed question in the Lecture 1 Notebook, I would like to have solid justification for the answer.

Does anybody have any verified justification on the choice between accuracy/loss of 2 models vs. 1 model?

Thanks in advance.