Question

我有一个数据集，需要将其从月度数据转换为季度数据。

这是我数据的前五行。

                   Measure Name             Year    Month   Value
0   Revenue from Sale of Recycled Materials 2007    Jan $1,757,000
1   Revenue from Sale of Recycled Materials 2007    Feb $2,052,000
2   Revenue from Sale of Recycled Materials 2007    Mar $2,747,000
3   Revenue from Sale of Recycled Materials 2007    Apr $2,308,000
4   Revenue from Sale of Recycled Materials 2007    May $2,289,000

我不知道从哪里开始将每月转换为每季度。

Jan-Mar will be Q1
April-June will be Q2
July-September will be Q3
October-December will be Q4.

我考虑创建Chort组分析，但是由于我的时间数据位于两列中，所以我不知道从哪里开始。

到目前为止，这是我的代码

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import matplotlib as mpl
%matplotlib inline
from pandas_datareader import data as pdr
from pandas.plotting import autocorrelation_plot
import seaborn as sns
from sklearn.metrics import accuracy_score, classification_report
plt.style.use(style='ggplot')

from datetime import datetime
from datetime import timedelta

recycle=pd.read_csv('Pathway link',
                       sep=',',)

我希望2007 Jan-Mar是Q1，依此类推。...

Answer 1

您可以通过多种方式解决此问题，首先是将Months视为数字，并应用一些逻辑测试语句。例如if month <= 3 THEN QTR = 'Q1'

但是似乎您正在处理字符串而不是datetime对象，因此我们可以尝试将月份传递到dict中并将“ Quarter”作为值对。当然，这取决于您的月份与密钥相同。

months = { 'Jan' : 'Q1',
      'Feb' : 'Q1',
      'Mar' : 'Q1',
      'Apr' : 'Q2',
      'May' : 'Q2',
      'Jun' : 'Q2',
      'Jul' : 'Q3',
      'Aug' : 'Q3',
      'Sep' : 'Q3',
      'Oct' : 'Q4',
      'Nov' : 'Q4',
      'Dec' : 'Q4' }

df['Qtr'] = df['Month'].map(months)
print(df)
    Measure Name    Year    Month   Value   QTR
0   Revenue from Sale of Recycled Materials 2007    Jan $1,757,000  Q1
1   Revenue from Sale of Recycled Materials 2007    Feb $2,052,000  Q1
2   Revenue from Sale of Recycled Materials 2007    Mar $2,747,000  Q1
3   Revenue from Sale of Recycled Materials 2007    Apr $2,308,000  Q2
4   Revenue from Sale of Recycled Materials 2007    May $2,289,000  Q2

然后您可以通过汇总应用您的组：

df.groupby('QTR')['Value'].sum() 
#you'll need to convert your value into a number if its an object.
QTR
Q1    6556000
Q2    4597000
Name: Value, dtype: int64

如何将每月数据转换为季度数据

1 个答案: