Using more complicated aggregation functions with groupby

In the previous recipe, we created a groupby DataFrame object and used it to run summary statistics by groups. We use chaining in this recipe to create the groups, choose the aggregation variable(s), and select the aggregation function(s), all in one line. We also take advantage of the flexibility of the groupby object, which allows us to choose the aggregation columns and functions in a variety of ways.

Getting ready

We will work with the National Longitudinal Survey of Youth (NLS) data in this recipe.

Data note

The NLS, administered by the United States Bureau of Labor Statistics, are longitudinal surveys of individuals who were in high school in 1997 when the surveys started. Participants were surveyed each year through 2018. The surveys are available for public use at nlsinfo.org.

How to do it…

We do more complicated aggregations with groupby than we did in the previous recipe, taking advantage...