How to Accurately Convert Month Strings to End‑of‑Month Dates with Pandas

1. Introduction

In a Python community chat a user asked how to handle month‑level date strings in Pandas, as shown in the screenshot below.

2. Implementation Process

At first the idea was to simply append "31" to each month and format the string, but because months have different numbers of days this approach fails. A built‑in Pandas function can handle the conversion correctly.

Method One

Code:

import pandas as pd
from pandas.tseries.offsets import MonthEnd

df = pd.DataFrame({
    'date': ['2022-01', '2022-02', '2022-03', '2022-04', '2022-05', '2022-06', '2022-07']
})
df['new_date'] = df['date'].astype('datetime64').map(MonthEnd())
print(df)

Result:

Method Two

For compact strings such as "202201", convert them with pd.to_datetime using the appropriate format, then apply MonthEnd:

import pandas as pd
from pandas.tseries.offsets import MonthEnd

df = pd.DataFrame({
    'date': ['202201', '202202', '202203', '202204', '202205', '202206', '202207']
})
df['new_date'] = pd.to_datetime(df['date'], format='%Y%m').map(MonthEnd())
print(df)

Result:

During testing it was discovered that if the original date is already the last day of the month, the conversion may shift to the last day of the following month, causing a one‑day offset.

Dependency‑Free Approach

A solution that avoids extra imports and third‑party dependencies was also shared, illustrated in the image below.

3. Summary

The article presented a common Pandas time‑handling problem, explained why naive string manipulation fails, and provided two reliable code snippets—one using explicit month strings and another using compact numeric strings—along with a note on edge‑case behavior and a lightweight, dependency‑free alternative.