We have come a long way, but we're not quite done yet. We need to talk about hierarchical indexing. In this section, we look at hierarchical indices, why they are useful, how they are created, and how they can be used.
So, what are hierarchical indices? They bring additional structure to an index and exist in pandas as MultiIndex class objects, but they are still an index that can be assigned to a series or DataFrame. With a hierarchical index, we think of rows in a DataFrame, or elements in a series, as uniquely identified by combinations of two or more indices. These indices have a hierarchy, and selecting an index at one level will select all elements with that level of the index. We can go on a more theoretical path and claim that when we have a MultiIndex, the dimensionality of the table increases. It behaves, not as a square on which data exists, but as a cube, or at least it could.
A hierarchical index is used when we want additional structure on the index without...