I am trying to project out the value in one column by adding a fixed value. Essentially, it would be like a sheet in Excel where there would be a value in cell a1, a fixed value in cell b1, and a2 would be =sum(a1+$b$1) and dragging the cells down. It seems like that would be straightforward in pandas, perhaps involving .cumsum, but I haven't found a way to get my desired result.
Here is my data frame.
df = pd.DataFrame({"date_column" : ["2020-08-16", "2020-08-17", "2020-08-18", "2020-08-19", "2020-08-20" , "2020-08-21", "2020-08-22", "2020-08-23", "2020-08-24", "2020-08-25" ] ,
"purchases" : [23,'','','','','','','','','']})
df
date_column purchases
0 2020-08-16 23
1 2020-08-17
2 2020-08-18
3 2020-08-19
4 2020-08-20
5 2020-08-21
6 2020-08-22
7 2020-08-23
8 2020-08-24
9 2020-08-25
Separately, I have calculated a variable called avg_increase for the preceding two weeks.
avg_increase
2.5000000
I want to add the avg_increase value to the initial value in the purchases column, and then add the avg_increase to that value, and so on. The desired output is:
date_column purchases
0 2020-08-16 23
1 2020-08-17 25.5
2 2020-08-18 28
3 2020-08-19 30.5
4 2020-08-20 33
5 2020-08-21 35.5
6 2020-08-22 38
7 2020-08-23 40.5
8 2020-08-24 43
9 2020-08-25 45.5
Alternative solution assuming ordered index 0..N:
df.purchases = df.loc[0,"purchases"]+df.index.values*2.5
result:
date_column purchases
0 2020-08-16 23.0
1 2020-08-17 25.5
2 2020-08-18 28.0
3 2020-08-19 30.5
4 2020-08-20 33.0
5 2020-08-21 35.5
6 2020-08-22 38.0
7 2020-08-23 40.5
8 2020-08-24 43.0
9 2020-08-25 45.5
Simply get first value of purchases and add index values multiplied by 2.5