CODEWITHSUNDEEP
python-3.xpandasdataframe
user3889486
user3889486

Reputation: 656

Divide multiple columns in pandas DF

Given the df2 DF, where

df2[car_colname]
             car_30    car_90  ...
2005-02-21  0.117660  0.004845 ...
2005-02-22  0.145713  0.000901 ...
2005-02-23  0.074074  0.002139 ...
2005-02-24  0.076934 -0.004596 ...
2005-02-25  0.085200 -0.016227 ...
2005-04-08  0.097728 -0.005937 ...
2005-04-11  0.222366  0.007553 ...

and 

df2[sd_colname]
              std_30    std_90 ...
2005-02-21  0.052266  0.004150 ...
2005-02-22  0.052266  0.004150 ...
2005-02-23  0.052266  0.004150 ...
2005-02-24  0.052266  0.004150 ...
2005-02-25  0.052266  0.004150 ...
2005-04-08  0.061682  0.006731 ...
2005-04-11  0.061682  0.006731 ...

I want to divide row-wise df2[car_colname] by df2[sd_colname] so that the result in the first row comes from: 0.117660 / 0.052266 = 2.25117667317185 and 0.004845 / 0.004150 = 1.16746987951807, and so on...

2005-02-21      2.25117667317185    1.16746987951807   ...
2005-02-22      2.78791183561015    0.21710843373494   ...
2005-02-23      1.41725022002832    0.515421686746988  ...
2005-02-24      1.4719703057437    -1.10746987951807   ...
2005-02-25      1.6301228331994    -3.91012048192771   ...
2005-04-08      1.58438442333258   -0.882038330114396  ...
2005-04-11      3.60503874712234    1.12212152726192   ...

I've tried df2[car_colname] / df2[sd_colname] or df2[car_colname].div(df2[sd_colname]) to no avail.

Upvotes: 0

Views: 37

Answers (2)

Vaishali
Vaishali

Reputation: 38415

Just replace the columns names,

df1.div(df2.rename(columns = (dict(zip(df2.columns, df1.columns)))))


            car_30      car_90
2005-02-21  2.251177    1.167470
2005-02-22  2.787912    0.217108
2005-02-23  1.417250    0.515422
2005-02-24  1.471970    -1.107470
2005-02-25  1.630123    -3.910120
2005-04-08  1.584384    -0.882038
2005-04-11  3.605039    1.122122

Upvotes: 0

BENY
BENY

Reputation: 323226

You may need to change your columns make it all same in two dfs 

df1.columns=df1.columns.str.split('_').str[-1]
df2.columns=df2.columns.str.split('_').str[-1]
df1.div(df2)
Out[786]: 
                  30        90
2005-02-21  2.251177  1.167470
2005-02-22  2.787912  0.217108
2005-02-23  1.417250  0.515422
2005-02-24  1.471970 -1.107470
2005-02-25  1.630123 -3.910120
2005-04-08  1.584384 -0.882038
2005-04-11  3.605039  1.122122

Upvotes: 2

Related Questions