Combine groups after iteration

Question

I have a dataframe that contains values for countries and years:

country year value
US       2000  20
JP       2000  10
AU       2000   5
US       2001  22
JP       2001  12
AU       2001   6
US       2002  23
JP       2002  14
AU       2002   8

I want to calculate the percent change for each country between years, so I group by countries and iterate per group:

grouped=df.groupBy('country')
for group in grouped:
    group['pct']=group['value'].pct_change(periods=1)*100

How can I create a new dataframe from 'grouped' containing my new column pct?

Answer 1

Just put your code in a function and use apply:

def f(group):
    group['pct']=group['value'].pct_change(periods=1)*100
    return group    
print df.groupby('country').apply(f)

Output:

  country  year  value        pct
0      US  2000     20        NaN
1      JP  2000     10        NaN
2      AU  2000      5        NaN
3      US  2001     22  10.000000
4      JP  2001     12  20.000000
5      AU  2001      6  20.000000
6      US  2002     23   4.545455
7      JP  2002     14  16.666667
8      AU  2002      8  33.333333