One model per category
The model you ran on the whole dataset fits some parts of the data better than others. It's worth taking a look at what happens when you run a linear model on different parts of the dataset separately, to see if each model agrees or disagrees with the others.
taiwan_real_estate is available.
This exercise is part of the course
Intermediate Regression with statsmodels in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Filter for rows where house age is 0 to 15 years
taiwan_0_to_15 = ____
# Filter for rows where house age is 15 to 30 years
taiwan_15_to_30 = ____
# Filter for rows where house age is 30 to 45 years
taiwan_30_to_45 = ____