Variable Selection using RFE in Multiple Linear Regression with Python

Even though the housing dataset doesn’t have many variables and you can easily select the relevant features manually, it is important to learn to use certain automated techniques as well.

Let’s see Rahim rebuild another model for the housing dataset using RFE.

Before that please download the python notebook for model building using RFE given below. The dataset used in this case is the same dataset, i.e. the 'Housing' dataset that you had used earlier.

In order to use RFE, you need to use SKLearn instead of statsmodels. Let’s go ahead and create a linear model using SKLearn to perform RFE.

Now that the features have been selected and the model is built, let's do the residual analysis and make the final predictions.