Practical Predictive Analytics

Merging data using Inner and Outer joins

Now let's merge the membership file with the purchases. In SQL there are two kinds of merge that you can consider when associating two dataframes. An inner join will consolidate two records based on the matching of a single or multiple key that they both have in common. An outer join will also merge the two tables together by keys, but will also include any rows that are not matched. You can identify an outer join by observing the existence of NAs in any of the matching keys. Inner joins are usually more efficient, but should only be used if you expect to have matching keys of both of the joined files.

The join2 dataframe is an inner join that will contain only those members that had a purchase record. ...