Apply a function on a pandas column

Question

I have created the function below which returns a vector based on two input vectors. I have a pandas dataframe with different columns and I would like to apply that function to some columns of my dataframe (one pandas column would be the first vector of my function (nominal), another one the second vector of my function(CPI) and I would assign the result of my function to a new pandas column). How could I do it?

Thank you very much for your help,

Pierre

nominal=[10,10,10,10,10,10,10,10,10,10,10,10]
CPI=[0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02]

def nominal_to_real(nominal,CPI):
    real=[]
    product=1
    for i in range(len(nominal)):
        product*=(CPI[i]+1)**(1/12)     
        real.append(nominal[i]/product)
    return real

um, nominal_to_real(df.nominal, df.CPI)? But note, your function returns a list that has one more element than your input lists... so I'm not sure how you expect this to behave with a data-frame... — juanpa.arrivillaga
– juanpa.arrivillaga, Commented Mar 9, 2018 at 0:35
@juanpa.arrivillaga, I think (read hope) he's asking how to vectorize this, which is a fair question since there seems to be sum kind of cumproduct feature. — jpp
– jpp, Commented Mar 9, 2018 at 0:45
Give a Minimal, Complete, Verifiable Example in your question. — Alex
– Alex, Commented Mar 9, 2018 at 1:56
@juanpa.arrivillaga Thank you very much! That worked perfectly! My list has indeed on more element than my inputs, my mistake. I just needed to delete real.append(nominal[0]) and everything is perfect. — Peslier53
– Peslier53, Commented Mar 9, 2018 at 9:23

jezrael · Accepted Answer · 2018-03-13 12:55:35Z

You can use cumprod for vectorized solution:

nominal=[10,10,10,10,10,10,10,10,10,10,10,10]
CPI=[0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02,0.02]

def nominal_to_real(nominal,CPI):
    real=[]
    product=1
    for i in range(len(nominal)):
        product*=(CPI[i]+1)**(1/12)     
        real.append(nominal[i]/product)
    return real

a = nominal_to_real(nominal,CPI)
df = pd.DataFrame(np.c_[nominal, CPI, a], columns=['nominal','CPI','real'])

df['real1'] = df['nominal'] / ((df['CPI'] + 1)**(1/12)).cumprod() 
print (df)
    nominal   CPI      real     real1
0      10.0  0.02  9.983511  9.983511
1      10.0  0.02  9.967050  9.967050
2      10.0  0.02  9.950616  9.950616
3      10.0  0.02  9.934209  9.934209
4      10.0  0.02  9.917829  9.917829
5      10.0  0.02  9.901475  9.901475
6      10.0  0.02  9.885149  9.885149
7      10.0  0.02  9.868850  9.868850
8      10.0  0.02  9.852578  9.852578
9      10.0  0.02  9.836332  9.836332
10     10.0  0.02  9.820114  9.820114
11     10.0  0.02  9.803922  9.803922

Collectives™ on Stack Overflow

Apply a function on a pandas column

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related