Using groupby to organize data by groups

At a certain point in most data analysis projects, we have to generate summary statistics by groups. While this can be done using the approaches in the previous recipe, in most cases the pandas DataFrame groupby method is a better choice. If groupby can handle an aggregation task—and it usually can—it is likely the most efficient way to accomplish that task. We make good use of groupby in the next few recipes. We go over the basics in this recipe.

Getting ready

We will work with the COVID-19 daily data in this recipe.

How to do it…

We will create a pandas groupby DataFrame and use it to generate summary statistics by group:

1. Import pandas and numpy, and load the COVID-19 daily data:

   import pandas as pd
   coviddaily = pd.read_csv("data/coviddaily.csv", parse_dates=["casedate"])
   

2. Create a pandas groupby DataFrame:

   countrytots = coviddaily.groupby([&apos...