CODEWITHSUNDEEP
pythonpandasdataframedatetime
Somphon Rueangari
Somphon Rueangari

Reputation: 65

Percentage of each day where sum is the total of each day value

I am trying to get the percentage of each day where sum is given.

I have data in daily with datetime index and i resemble index to yearly using method sum and here is the code.

data_converted = data.resample('AS').sum()

and what I want is to get the percentage of each day:

Here is what I wanted to do

percentage=[ (data[all_day_in_year2000] / data_converted[year2000] )*100 ]

Noted that my data has more than one year

If you think i am asking too much than you can neglect this one:

It would be nice if it's flexible incase I resemble daily data to monthly data or quarter data.

Thank you for your consideration.

Edit

Sorry for not posting my data

Here is my data.

1986-01-02                                              25.56  
1986-01-03                                              26.00  
1986-01-06                                              26.53  
1986-01-07                                              25.85  
1986-01-08                                              25.87  
1986-01-09                                              26.03  
1986-01-10                                              25.65  
1986-01-13                                              25.08  
...                                                       ...  
2019-06-04                                              53.50  
2019-06-05                                              51.57  
2019-06-06                                              52.59  
2019-06-07                                              53.95  
2019-06-10                                              53.33  
2019-06-11                                              53.30

When i resample using sum i got the following.

1986-01-01     3776.97
1987-01-01     4876.93
1988-01-01     4103.11
1989-01-01     5046.32
1990-01-01     6303.33
1991-01-01     5514.59

and what i want is the percent of the first one respective to second.It can also be store in list with just the percentage alone without datetime:

                    #value in daily /value in yearly*100
1986-01-02          (25.56 / 3776.97)*100
1986-01-03          (26.00 / 3776.97)*100
1986-01-06          (26.53 / 3776.97)*100
1986-01-07          (25.85 / 3776.97)*100

similarly for next year:

1986-02-03          (17.42 / 4876.93)*100

Upvotes: 1

Views: 287

Answers (3)

BhishanPoudel
BhishanPoudel

Reputation: 17164

Just replace month by year in this answer, you are good to go:

import numpy as np
import pandas as pd

df = pd.DataFrame({'a': pd.date_range('1986-01-01','1986-03-01',closed='left')})
df['b'] = df.a.dt.month

# this is the answer, just one liner
df['answer']=df.b/df.groupby(df.a.dt.month).b.transform('sum')*100

# just looking the sum, this is not required
df['just_check_sum'] = df.groupby(df.a.dt.month).b.transform('sum')

print(df.head())
           a  b    answer  just_check_sum
0 1986-01-01  1  3.225806              31 ===> 1/31*100 = 3.22 %
1 1986-01-02  1  3.225806              31
2 1986-01-03  1  3.225806              31
3 1986-01-04  1  3.225806              31
4 1986-01-05  1  3.225806              31

Upvotes: 0

Teemu Kurppa
Teemu Kurppa

Reputation: 4849

I prefer to use groupby and then apply per group: 

by_year = data.groupby(pd.Grouper(freq='AS')) 
percentage = by_year.apply(lambda x: x/x.sum() * 100)

Upvotes: 2

Andy L.
Andy L.

Reputation: 25259

you may use transform with resample and doing calculation between daily values with result from transform of sum

Your Sample data: (I just grab partial data which you posted)

Out[11]:
              val
dates
1986-01-02  25.56
1986-01-03  26.00
1986-01-06  26.53
1986-01-07  25.85
1986-01-08  25.87
1986-01-09  26.03
1986-01-10  25.65
1986-01-13  25.08
2019-06-04  53.50
2019-06-05  51.57
2019-06-06  52.59
2019-06-07  53.95
2019-06-10  53.33
2019-06-11  53.30

df.val / df.resample('AS').val.transform('sum') * 100

Out[14]:
dates
1986-01-02    12.373530
1986-01-03    12.586532
1986-01-06    12.843104
1986-01-07    12.513918
1986-01-08    12.523600
1986-01-09    12.601055
1986-01-10    12.417098
1986-01-13    12.141163
2019-06-04    16.811212
2019-06-05    16.204751
2019-06-06    16.525264
2019-06-07    16.952614
2019-06-10    16.757793
2019-06-11    16.748366
Name: val, dtype: float64

Upvotes: 3

Related Questions