In the previous section, it was observed that predictions from different models tend to diverge more significantly at the edges of the data.

To be more precise, the predictions start to exhibit unusual behavior when we go beyond the range of values present in the training set.
Predicting values outside the range of the training set is referred to as extrapolation, while predicting values within the range is called interpolation.

Regression models are not well-suited for handling extrapolation.
They are primarily used for interpolation and may produce unreliable or nonsensical predictions when new instances fall outside the range of the training set.