I got Column1, Column2, Column3, Column4, Column5, Column6

I'd like to group Column1 and get the row sum of Column3,4 and 5

When I apply groupby() and get this that is correct but it's leaving out Column6:

    df = df.groupby(['Column1'])[['Column3', 'Column4', 'Column5']].sum

I tried with this but it doesn't group according to Column1 and it doesn't sum anything, but I get all my columns:

    df.sort_values(['Column1']).groupby(['Column3', 'Column4', 'Column5']).sum()

How can I use groupby() correctly in this case?

Thank you!

I add my code:

    df = df.drop(['Position', 'Swap', 'S / L', 'T / P'], axis=1)
    df = df.groupby(['Symbol']).agg({'Profit': ['sum'], 'Volume': ['sum'], 'Commission': ['sum'], 'Time': pd.Series.mode})
    df['Comm. ratio'] = (df['Commission'] / df['Profit'])
    df['Comm. ratio'] = df['Comm. ratio'].round(2)
    df['Net profit'] = (df['Profit'] + df['Commission'])
    df.loc['Total'] = pd.Series([df['Commission'].sum(),df['Profit'].sum(),df['Net profit'].sum()], index = ['Commission','Profit','Net profit'])

The output is:


As you can see it adds "sum" and "mode" rows that I'd like not to have.

Moreover, it ignores the df.loc['Total'] code and it leaves the Total row empty.

  • $\begingroup$ I added the changes you can make to fix the column names and to add the overall totals to my answer below. $\endgroup$
    – Donald S
    Commented Jul 11, 2020 at 9:04

1 Answer 1


I think the answer depends on what you want to do with column 6. Keep in mind that the values for column6 may be different for each groupby on columns 3,4 and 5, so you will need to decide which value to display. Typically, when using a groupby, you need to include all columns that you want to be included in the result, in either the groupby part or the statistics part of the query.

If you don't want to group by that column, you can just display the min or mode value. In general, if you want to calculate statistics on some columns and keep multiple non-grouped columns in your output, you can use the agg function within the groupyby function.

Example with most common value for column6 displayed:

df.groupby('Column1').agg({'Column3': ['sum'], 'Column4': ['sum'], 'Column5': ['sum'], 'Column6': pd.Series.mode})

Full example with code:

enter image description here

If there is a tie for most common, with one Mary and one Jane both being Female Engineers, this will generate an error as mode doesn't reduce to a single answer:

enter image description here

You will need to use another aggregate in that case, such as min, which will choose Jane as an alphanumeric min:

enter image description here

If you don't like the look of the multi-index, you can remove it using as_index=False and replacing the column names with a list(map(...join...))

enter image description here

or remove it using to_flat_index() which gives a slightly different format for the columns:

enter image description here

  • $\begingroup$ Thanks for the suggestion. It seems working somehow but I don't understand the pd.Series.mode to keep other columns. It indeed works but it adds another row with "sum" and "mode". Also how it is possible to include more than 1 column along with Column6, such as Column7, Column8. And why to sum different columns is not possible to group them in a single .sum(), rather than selecting 1 colum and sum it, and so on? I'll post my code below to explain better the situation. $\endgroup$
    – Steven
    Commented Jul 11, 2020 at 7:53
  • $\begingroup$ mode is a also a group by function. It just selects the most common value given the grouping. For example, in the table from my code, you can see the name "Joe" 2 times, so it is the most common value for M and Engineer. This is why it shows up in the mode column. $\endgroup$
    – Donald S
    Commented Jul 11, 2020 at 8:09
  • 1
    $\begingroup$ I added some examples above on how to remove the extra row/multi-index with "sum" and "mode". You can sum multiple columns into one column as a 2nd step by adding a new column as a sum of sums column, df['total_sum'] = df['column3sum'] + df['column4sum'] etc. $\endgroup$
    – Donald S
    Commented Jul 11, 2020 at 8:40
  • 1
    $\begingroup$ It may be having trouble with determining the mode from the data. Try this instead: df.groupby('Column1').agg({'Column3': ['sum'], 'Column4': ['sum'], 'Column5': ['sum'], 'Column6': 'min', 'Column7': 'min', 'Column8': 'min'}) $\endgroup$
    – Donald S
    Commented Jul 11, 2020 at 9:43
  • 1
    $\begingroup$ Yes, it worked. It's tricky how to figure this things out. $\endgroup$
    – Steven
    Commented Jul 11, 2020 at 9:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.