The bottom-up approach was just to give you an idea of how the parameters change when the number of variables is increased. More generally, you first build a model using all and then try to improve the model by dropping some of them.

Letâ€™s now build a model with all the available variables.

In the following video, you will use two main parameters to judge the insignificant variables, the p-values and the VIFs.

## Coming up

Now that we seem to have a fair model, in the next segment, let’s perform the final steps – **analyse the residual terms** and **predict** the price.