Summing up previous 10 rows of a dataframe

Question

I'm wondering how to sum up 10 rows of a data frame from any point.

I tried using rolling(10,window =1).sum() but the very first row should sum up the 10 rows below. Similar issue with cumsum()

So if my data frame is just the A column, id like it to output B.

    A   B
0   10  550
1   20  650
2   30  750
3   40  850
4   50  950
5   60  1050
6   70  1150
7   80  1250
8   90  1350
9   100 1450
10  110 etc
11  120 etc
12  130 etc
13  140 
14  150 
15  160 
16  170 
17  180 
18  190 

It would be similar to doing this operation in excel and copying it down

Excel Example:

Answer 1

You can reverse your series before using pd.Series.rolling, and then reverse the result:

df['B'] = df['A'][::-1].rolling(10, min_periods=0).sum()[::-1]

print(df)

      A       B
0    10   550.0
1    20   650.0
2    30   750.0
3    40   850.0
4    50   950.0
5    60  1050.0
6    70  1150.0
7    80  1250.0
8    90  1350.0
9   100  1450.0
10  110  1350.0
11  120  1240.0
12  130  1120.0
13  140   990.0
14  150   850.0
15  160   700.0
16  170   540.0
17  180   370.0
18  190   190.0