How to Quickly Get Month Start and End Dates with Pandas

Data Requirement

The example uses an online retail dataset (Online_Retail.csv.zip) with an InvoiceDate column parsed as datetime.

import pandas as pd
# Parse InvoiceDate as datetime
df = pd.read_csv('Online_Retail.csv.zip', parse_dates=['InvoiceDate'])
df = df.dropna().copy()

Solution Approaches

1. Manual datetime construction

Separate year and month, then create a new date with day set to 1.

def get_month_start(x):
    return datetime(x.year, x.month, 1)

df['MonthStart'] = df['InvoiceDate'].map(get_month_start)

2. pandas offsets: MonthBegin and MonthEnd

pandas provides built‑in offset classes to obtain month boundaries.

from pandas.tseries.offsets import MonthBegin, MonthEnd

# Sample dates for demonstration
df2 = pd.to_datetime([
    '2022-9-1','2022-9-2','2022-9-29','2022-9-30',
    '2022-10-1','2022-10-2','2022-10-30','2022-10-31'
]).to_frame(name='date')

Setting the offset parameter n to 0 returns the month start/end for the current month; setting it to 1 shifts to the next month.

3. Using .dt.floor('D') with MonthBegin

When InvoiceDate includes a time component, truncate to the date first.

df['InvoiceDate'].dt.floor('D') + MonthBegin() - MonthBegin()

4. Period conversion ( period('M') )

Convert dates to a monthly period and then extract start or end dates.

# Create a monthly period range
df3 = pd.period_range('2021-10','2022-05',freq='M').to_frame(name='date')

# Convert existing datetime series to period
df['InvoiceDate'].dt.to_period('M')

Period objects expose start_time and end_time. The end_time includes the final millisecond, so apply .floor('D') to obtain a clean date.

5. Using .dt.asfreq

Change the frequency of a period to daily, specifying start ( 'S') or end ( 'E') of the month.

df3['date'].dt.asfreq('D', how='S')  # month start
df3['date'].dt.asfreq('D', 'E')      # month end

Performance tests show asfreq is slightly faster than accessing start_time directly, though the resulting types differ; convert back to timestamps with .to_timestamp() if needed.

Notes

If the date column is a string Series, convert it first with pd.to_datetime(s, format=...).

The source data can be downloaded from the link provided at the beginning of the article.

Conclusion

The article compares several pandas‑based methods for obtaining month‑start and month‑end dates, emphasizing that vectorized operations on pandas datetime objects are generally more concise and efficient than manual datetime constructions.