Output of pandas groupby with cumprod not showing groupby columns

Question

I am trying to understand the difference between a mean,sum function vs a cumprod function.
When I run a groupby and then the mean, I get the id column and a mean of the values as expected.

When I run it with cumprod though, there is no groupby column. How do I ensure that I can get the columns I am grouping by

x = [.25,.23,.55,.89,-.90,-.04]
id = ['a', 'a', 'a', 'b', 'b', 'b']
df.groupby('id').mean()
df.groupby('id').cumprod()

Answer 1

df.groupby('id').mean() is shorthand for df.groupby('id').agg('mean').

df.groupby('id').cumprod() is shorthand for df.groupby('id').transform('cumprod').

The key difference here is that the former is a groupby/agg operation, while the latter is a groupby/transform operation.

groupby/agg aggregates each group into a single value. Therefore, the groupby/agg operation can return a Series whose index contains groupby keys (in this case, id values).

groupby/transform operations return a Series with the same number rows as the original DataFrame, df. (cumprod is short for cumulative product. Since it returns a running total of the products, there is one value for each row). Since there is a value for each row of the original DataFrame, naturally the index can not be the groupby keys. It has to remain being the index of the original DataFrame.