how to sum two dataframes python

Question

I have two dataframes

tickers                        dt  AAPL  AMC  AMZN  ...  TH  TSLA  VIAC  WKHS
0       2021-03-22 00:00:00+00:00     0    0     0  ...   0     1     0     0
1       2021-03-23 00:00:00+00:00     0    0     0  ...   0     1     0     0
2       2021-03-24 00:00:00+00:00     1    2     0  ...   0     0     0     0
3       2021-03-25 00:00:00+00:00     0    0     0  ...   0     0     0     0
4       2021-03-26 00:00:00+00:00     0    2     0  ...   0     4     0     0

tickers                        dt  AAPL  AMC  AMZN  ...  TH  TSLA  VIAC  WKHS
0       2021-03-19 00:00:00+00:00     0    0     0  ...   0     0     0     0
1       2021-03-20 00:00:00+00:00     0    0     0  ...   0     0     0     0
2       2021-03-21 00:00:00+00:00     0    0     0  ...   0     0     0     0
3       2021-03-22 00:00:00+00:00     0    0     0  ...   0     3     0     0
4       2021-03-23 00:00:00+00:00     0    0     0  ...   0     3     0     0

I want to sum each row by corresponding row from another dataframe. You can see that some time from one dataframe doesn't exist in another dataframe. I also want to consider them and include in the new dataframe

Try this: https://stackoverflow.com/a/23361783/2612429 – S2L Aug 17 '21 at 09:40 — S2L, Aug 17 '21 at 09:40
yeah but I also want to consider when they don't intersect – YanRemes Aug 17 '21 at 09:47 — YanRemes, Aug 17 '21 at 09:47

score 3 · Accepted Answer · answered Aug 17 '21 at 09:52

3

Try this:

df = pd.concat([df1, df2]).groupby(['dt']).sum().reset_index()

print(df)

PS: This is ensure all datetimes to exist.

answered Aug 17 '21 at 09:52

Deepak Tatyaji Ahire

4,883
2
13
35

also, how would I divide one data by another keeping all the time rows – YanRemes Aug 17 '21 at 10:15
I request you to kindly ask a separate question as it will involve invalid division operation if the values are missing in either of the data frame. – Deepak Tatyaji Ahire Aug 17 '21 at 10:20

Ayush Goel · Answer 2 · 2021-08-17T11:03:46.013

2

One way would be to first merge the dataframes, and then carry out the sum.

df1 = df1.merge(df2, left_on='dt', right_on='dt', how='outer')

The column names may end with an '_x' or '_y' to allow you to differentiate. You can then do a normal dataframe sum.

edited Aug 17 '21 at 11:03

answered Aug 17 '21 at 09:41

Ayush Goel

365
1
7

I tried this one but it remove row with 2021-03-19 time. And I want to keep it as well – YanRemes Aug 17 '21 at 09:46
In that case, I think you can just set `how='outer'`. This will take the union of both keys, so they won't be removed. – Ayush Goel Aug 17 '21 at 11:03
the dataset doesn't have 2021-03-19 00:00:00+00:00 1 2021-03-20 00:00:00+00:00 2 2021-03-21 00:00:00+00:00 – YanRemes Aug 17 '21 at 11:14

how to sum two dataframes python

2 Answers2