27. Text: Recap

Recap

1. You learned how to build a multiple linear regression model in Python, which was actually very similar to what you did in the last lesson on simple linear regression.
   
   

2. You learned how to encode dummy variables, and interpret the coefficients attached to each.
   
   

3. You learned about higher order terms, and how this impacts your ability to interpret coefficients.
   
   

4. You learned how to identify what it would mean for an interaction to be needed in a multiple linear regression model, as well as how to identify other higher order terms. But again, these do make interpreting coefficients directly less of a priority, and move your model towards one that, rather, aims to predict better at the expense of interpretation.
   
   

5. You learned about the model assumptions, and we took a closer look at multicollinearity. You learned about variance inflation factors, and how multicollinearity impacts the model coefficients and standard errors.