Pivoting our data
As you saw, there does seem to be an increase in the number of purchases by purchasing users within their first week. Let's now confirm that this is not driven only by one segment of users. We'll do this by first pivoting our data by 'country'
and then by 'device'
. Our change is designed to impact all of these groups equally.
The user_purchases
data from before has been grouped and aggregated by the 'country'
and 'device'
columns. These objects are available in your workspace as user_purchases_country
and user_purchases_device
.
As a reminder, .pivot_table()
has the following signature:
pd.pivot_table(data, values, columns, index)
This exercise is part of the course
Customer Analytics and A/B Testing in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Pivot the data
country_pivot = pd.pivot_table(user_purchases_country, values=['____'], columns=['____'], index=['____'])
print(country_pivot.head())