CODEWITHSUNDEEP
pythonpandas
user308827
user308827

Reputation: 21961

computing weighted average in pandas dataframe

In the foll. dataframe:

                              Potential     Area       weight  
NAME                                                                    
Ket                              21.694198   61730.10  0.028773  
Chles                            31.023133  101288.07  0.047211  
Caline                           23.066401   68515.47  0.031936  
Algany                           22.794050  109111.05  0.050858  
Baire                            3.176830   20763.54  0.009678

How can I add a new row that for each column contains the sum of the product of that column with the weight column? I tried this:

df.loc['new_row'] = df.mean() * df['weight']

Upvotes: 1

Views: 62

Answers (1)

jpp
jpp

Reputation: 164623

Here's one way using NumPy:

# calculate sum of products
sums = (df.iloc[:, :2].values * df['weight'].values[:, None]).sum(0).tolist()

# assign to index label
df.loc['MeanWeightProd'] = sums + [np.nan]

print(df)

                Potential           Area    weight
NAME                                              
Ket             21.694198   61730.100000  0.028773
Chles           31.023133  101288.070000  0.047211
Caline          23.066401   68515.470000  0.031936
Algany          22.794050  109111.050000  0.050858
Baire            3.176830   20763.540000  0.009678
MeanWeightProd   4.015494   14496.300611       NaN

Upvotes: 1

Related Questions