CODEWITHSUNDEEP
pythonpandas
qfd
qfd

Reputation: 788

groupby difference and percentage change

I have the following data frame

              t_f      m_s
   uni        
   A          False    1.2
   A          True     0.6
   B          False    0.9
   B          True     0.6

I am trying to get the difference of True - False for each uni and also calculate the percentage change True-False/False I am sure there is a way to do this using group by but i cant figure out how.

The output should be a data frame

              m_s_diff   m_s_diff_percentage
   uni        
   A          -0.6         -50%
   B          -0.3         -33.33%

Upvotes: 0

Views: 476

Answers (2)

BENY
BENY

Reputation: 323226

Try this ..

df.groupby(level=0)['m_s'].diff().dropna()


uni
A   -0.6
B   -0.3
Name: m_s, dtype: float64

Below are solution provided by @MaxU 

df.groupby(level=0)['m_s'].agg(['diff','pct_change']).dropna()
Out[502]: 
     diff  pct_change
uni                  
A    -0.6   -0.500000
B    -0.3   -0.333333

Upvotes: 2

user2341726
user2341726

Reputation: 392

Use query.

df.query('t_f==True').m_s - df.query('t_f==False').m_s

Upvotes: 1

Related Questions