I am trying to Convert yy-mm-dd to Q1 , Q2 for each company ID.

faiza.zafar

Hi community,

I am working with institutional ownership data and I extracting quarterly historical investor data. I want to convert my dates for each company to quarterly format as is shown in the attached picture below.

My current data looks like this:

But I want it in the format as shown

I have had a look at @Alan Tam123 post https://community.developers.refinitiv.com/questions/60189/date-format-and-column-headings.html but the code seems to be different from how I am doing it.

Any help would be great! @Jirapongse look forward to your expert help.

My code is attached below:

df,e = ek.get_data(instruments=['AAPL.O','AAL.L'], fields=[
    'TR.CommonName',
    'TR.InvestorFullName("TheInvestorType":"113,108,107")',
    'TR.SharesHeld.calcdate',
    'TR.SharesHeld',
    'TR.PctOfSharesOutHeld',
    'SUM(TR.PctOfSharesOutHeld)',
    'TR.InvestorType',
    'TR.InvestorTypeId'],parameters={'SDate':'0','EDate':'-4', 'Period':'FQ0','Frq':'FQ'})
df
#EDate is the number of quaters 

df1= df
df1['id']=df1.groupby(['Instrument']).ngroup()
df1.set_index(['id','Calc Date'])

Jirapongse

@faiza.zafar

I am not a Python Dataframe expert so I used the for loop to check and change the values.

First, I added the Quarter and QuaterTemp columns to the data frame.

df,e = ek.get_data(instruments=['AAPL.O','AAL.L'], fields=[
    'TR.CommonName',
    'TR.InvestorFullName("TheInvestorType":"113,108,107")',
    'TR.SharesHeld.calcdate',
    'TR.SharesHeld',
    'TR.PctOfSharesOutHeld',
    'SUM(TR.PctOfSharesOutHeld)',
    'TR.InvestorType',
    'TR.InvestorTypeId'],parameters={'SDate':'0','EDate':'-4', 'Period':'FQ0','Frq':'FQ'})
df['Quarter'] = pd.PeriodIndex(df['Calc Date'], freq='Q')
df['QuarterTemp'] = df['Quarter']
df

Then, I used a for loop to change the values in the Quarter column according to the following conditions. Next, I dropped the QuarterTemp column.

for i in range(1, len(df)):
    if ((df.loc[i-1,'QuarterTemp'] == df.loc[i,'QuarterTemp']) and (df.loc[i-1,'Instrument'] == df.loc[i,'Instrument'])):
        df.loc[i, 'Quarter'] = ""
df = df.drop('QuarterTemp', axis=1)
df

Finally, I called the groupby and set_index methods.

df1= df
df1['id']=df1.groupby(['Instrument']).ngroup()
df1 = df1.set_index(['id','Quarter'])
df1