Python Pandas PeriodIndex - Convert the PeriodArray to the specified frequency

To convert the PeriodArray to a specified frequency, use the periodIndex.asfreq() method. This method allows you to change the frequency of periods while preserving the time information.

Syntax

PeriodIndex.asfreq(freq, how='end')

Parameters

• freq: The target frequency (e.g., 'M' for monthly, 'D' for daily)
• how: How to handle conversion ('start' or 'end')

Creating a PeriodIndex

First, let's create a PeriodIndex object with yearly frequency ?

import pandas as pd

# Create a PeriodIndex object with yearly frequency
periodIndex = pd.PeriodIndex(['2021-09-25 07:30:35', '2019-10-30 04:15:45',
                              '2020-07-15 02:55:15', '2022-06-25 09:40:55'], freq="Y")

print("Original PeriodIndex:\n", periodIndex)
print("\nFrequency:", periodIndex.freqstr)

Original PeriodIndex:
 PeriodIndex(['2021', '2019', '2020', '2022'], dtype='period[A-DEC]')

Frequency: A-DEC

Converting to Monthly Frequency

Convert the yearly periods to monthly frequency using asfreq() ?

import pandas as pd

periodIndex = pd.PeriodIndex(['2021-09-25 07:30:35', '2019-10-30 04:15:45',
                              '2020-07-15 02:55:15', '2022-06-25 09:40:55'], freq="Y")

# Convert to monthly frequency
monthly_periods = periodIndex.asfreq('M')
print("Converted to Monthly:\n", monthly_periods)

Converted to Monthly:
 PeriodIndex(['2021-12', '2019-12', '2020-12', '2022-12'], dtype='period[M]')

Converting to Different Frequencies

You can convert to various frequencies like daily, quarterly, or weekly ?

import pandas as pd

periodIndex = pd.PeriodIndex(['2021', '2020', '2019'], freq="Y")

print("Original (Yearly):", periodIndex)
print("To Quarterly:", periodIndex.asfreq('Q'))
print("To Monthly:", periodIndex.asfreq('M'))
print("To Daily:", periodIndex.asfreq('D'))

Original (Yearly): PeriodIndex(['2021', '2020', '2019'], dtype='period[A-DEC]')
To Quarterly: PeriodIndex(['2021Q4', '2020Q4', '2019Q4'], dtype='period[Q-DEC]')
To Monthly: PeriodIndex(['2021-12', '2020-12', '2019-12'], dtype='period[M]')
To Daily: PeriodIndex(['2021-12-31', '2020-12-31', '2019-12-31'], dtype='period[D]')

Using the 'how' Parameter

The how parameter controls whether to use the start or end of the period ?

import pandas as pd

periodIndex = pd.PeriodIndex(['2021'], freq="Y")

print("Original:", periodIndex)
print("End of year (default):", periodIndex.asfreq('D', how='end'))
print("Start of year:", periodIndex.asfreq('D', how='start'))

Original: PeriodIndex(['2021'], dtype='period[A-DEC]')
End of year (default): PeriodIndex(['2021-12-31'], dtype='period[D]')
Start of year: PeriodIndex(['2021-01-01'], dtype='period[D]')

Conclusion

The asfreq() method is essential for converting PeriodIndex objects between different time frequencies. Use the how parameter to control whether conversion uses the start or end of the period.