How to add a column to a grouped by DataFrame from the original DataFrame - Python

Question

I have this DataFrame:

                       Name    Year Publisher  Global_Sales
0                Wii Sports  2006.0  Nintendo         82.74
1         Super Mario Bros.  1985.0  Nintendo         40.24
2            Mario Kart Wii  2008.0  Nintendo         35.82
3         Wii Sports Resort  2009.0  Nintendo         33.00
4  Pokemon Red/Pokemon Blue  1996.0  Nintendo         31.37

I want to group it by Year and see max Global_Sales per Year:

comp_group=df_comparation.groupby('Year')['Global_Sales'].max()

I obtain:

Year
1980.0     4.31
1981.0     4.50
1982.0     7.81
1983.0     3.20
1984.0    28.31
1985.0    40.24
1986.0     6.51
1987.0     4.38
1988.0    17.28
1989.0    30.26
1990.0    20.61

Now I want to know what Publisher made the max Global_Sales and add it as a column:

Year     Global_Sales      Publisher
1980.0     4.31            Nintendo
1981.0     4.50            EA Sports
1982.0     7.81              ...
1983.0     3.20              ...
1984.0    28.31              ...  
1985.0    40.24              ...
1986.0     6.51              ...  
1987.0     4.38              ...
1988.0    17.28              ...
1989.0    30.26              ...
1990.0    20.61              ...

Thanks!

Answer 1

you can aggregate with .idxmax() instead to get the index of the maximal sale per year, and then index with it to get the result:

indexes = df.groupby("Year")["Global_Sales"].idxmax()
result  = df.loc[indexes, ["Year", "Global_Sales", "Publisher"]]

Answer 2

Group the dataframe by Year then apply a function to get the Global_Sales and Publisher for maximum Global_Sales:

(df
 .groupby('Year')
 .apply(lambda x: x.loc[x['Global_Sales'].idxmax(), ['Global_Sales', 'Publisher']])
 )

        Global_Sales Publisher
Year                          
1985.0         40.24  Nintendo
1996.0         31.37  Nintendo
2006.0         82.74  Nintendo
2008.0         35.82  Nintendo
2009.0         33.00  Nintendo