Pandas date_range(): Generate Date & Time Ranges

pandas.date_range() generates a sequence of dates or timestamps and returns them as a DatetimeIndex. We can control the starting date, ending date, number of periods, frequency, timezone and whether the boundary dates are included.

Important: Use pd.date_range(), not DataFrame.date_range(). The function belongs directly to Pandas and returns a DatetimeIndex.

Show Table of Contents

Syntax of pandas.date_range() 🔝

pd.date_range(
    start=None,
    end=None,
    periods=None,
    freq=None,
    tz=None,
    normalize=False,
    name=None,
    inclusive='both',
    unit=None
)

Parameter	Purpose
`start`	Starting date or timestamp.
`end`	Ending date or timestamp.
`periods`	Number of timestamps to generate.
`freq`	Frequency such as day, hour, month start or month end.
`tz`	Timezone of the resulting DatetimeIndex.
`normalize`	Normalize start and end timestamps to midnight before generating the range.
`name`	Name assigned to the resulting DatetimeIndex.
`inclusive`	Control whether start and end boundaries are included.
`unit`	Set the datetime resolution such as seconds, milliseconds, microseconds or nanoseconds.

Common valid combinations are:

start + end
start + periods
end + periods
start + end + periods
start + periods + freq
end + periods + freq
start + end + freq

When start and end are supplied without periods, the usual default frequency is one calendar day.

Basic Date Range Using start and end 🔝

import pandas as pd

dates=pd.date_range(
    start='2020-04-20',
    end='2020-04-27'
)

print(dates)

Output

DatetimeIndex(['2020-04-20', '2020-04-21', '2020-04-22', '2020-04-23',
               '2020-04-24', '2020-04-25', '2020-04-26', '2020-04-27'],
              dtype='datetime64[us]', freq='D')

The default frequency is daily, represented by D. Both the starting and ending dates are included by default.

The displayed datetime resolution such as datetime64[us] or datetime64[ns] can differ between Pandas versions and depending on the supplied values. The generated dates remain the important part of the result.

Generate a Fixed Number of Dates with periods 🔝

Use periods to specify how many timestamps should be generated.

start + periods

import pandas as pd

dates=pd.date_range(
    start='2026-09-01',
    periods=5
)

print(dates)

Output

DatetimeIndex(['2026-09-01', '2026-09-02', '2026-09-03',
               '2026-09-04', '2026-09-05'],
              dtype='datetime64[us]', freq='D')

end + periods

dates=pd.date_range(
    end='2026-09-05',
    periods=5
)

This generates five daily dates ending on 5 September 2026.

start + end + periods

If start, end and periods are supplied without a frequency, Pandas creates evenly spaced timestamps between the two boundaries.

dates=pd.date_range(
    start='2020-04-20',
    end='2020-04-27',
    periods=3
)

print(dates)

Output

DatetimeIndex(['2020-04-20 00:00:00',
               '2020-04-23 12:00:00',
               '2020-04-27 00:00:00'],
              dtype='datetime64[us]', freq=None)

The timestamps are evenly distributed between the start and end values. They are not automatically restricted to complete days.

Generate Dates with freq 🔝

The freq parameter determines the distance between consecutive timestamps.

dates=pd.date_range(
    start='2026-09-01',
    periods=5,
    freq='D'
)

Common frequency aliases include:

Alias	Frequency
`D`	Calendar day
`B`	Business day
`W`	Week
`MS`	Month start
`ME`	Month end
`QS`	Quarter start
`QE`	Quarter end
`YS`	Year start
`YE`	Year end
`h`	Hour
`min`	Minute
`s`	Second

See the Pandas frequency aliases section for more date and time frequencies.

Daily and Custom Day Frequencies 🔝

Every Day

dates=pd.date_range(
    start='2026-09-01',
    periods=5,
    freq='D'
)

Every Three Days

dates=pd.date_range(
    start='2026-09-01',
    periods=5,
    freq='3D'
)

print(dates)

Output

DatetimeIndex(['2026-09-01', '2026-09-04', '2026-09-07',
               '2026-09-10', '2026-09-13'],
              dtype='datetime64[us]', freq='3D')

Combine Days and Hours

dates=pd.date_range(
    start='2026-09-01',
    periods=4,
    freq='3D6h'
)

Each timestamp is 3 days and 6 hours after the previous one.

Generate Business Days 🔝

Use B to generate standard business days. Saturdays and Sundays are skipped.

import pandas as pd

dates=pd.date_range(
    start='2026-09-04',
    periods=5,
    freq='B'
)

print(dates)

Output

DatetimeIndex(['2026-09-04', '2026-09-07', '2026-09-08',
               '2026-09-09', '2026-09-10'],
              dtype='datetime64[us]', freq='B')

This standard business-day frequency excludes weekends. It does not automatically know every country's public holidays.

Month Start and Month End 🔝

Month Start with MS

dates=pd.date_range(
    start='2026-01-01',
    periods=6,
    freq='MS'
)

print(dates)

Output

DatetimeIndex(['2026-01-01', '2026-02-01', '2026-03-01',
               '2026-04-01', '2026-05-01', '2026-06-01'],
              dtype='datetime64[us]', freq='MS')

Month End with ME

dates=pd.date_range(
    start='2026-01-01',
    periods=6,
    freq='ME'
)

print(dates)

Output

DatetimeIndex(['2026-01-31', '2026-02-28', '2026-03-31',
               '2026-04-30', '2026-05-31', '2026-06-30'],
              dtype='datetime64[us]', freq='ME')

Current Pandas: Use ME for month end. Older tutorials may show M, but the old month-end alias was removed in Pandas 3.0.

Year-End Frequency 🔝

Use YE for year end. An ending month can also be specified.

Year Ending in February

dates=pd.date_range(
    start='2010-04-20',
    end='2020-04-27',
    freq='YE-FEB'
)

print(dates)

Output

DatetimeIndex(['2011-02-28', '2012-02-29', '2013-02-28',
               '2014-02-28', '2015-02-28', '2016-02-29',
               '2017-02-28', '2018-02-28', '2019-02-28',
               '2020-02-29'],
              dtype='datetime64[us]', freq='YE-FEB')

The leap-year dates show 29 February where applicable.

Older Pandas examples may use A-FEB. Current Pandas uses the year-end alias YE-FEB.

Generate Hours, Minutes and Seconds 🔝

Hourly Range

dates=pd.date_range(
    start='2026-09-05 09:00',
    periods=5,
    freq='h'
)

print(dates)

Output

DatetimeIndex(['2026-09-05 09:00:00', '2026-09-05 10:00:00',
               '2026-09-05 11:00:00', '2026-09-05 12:00:00',
               '2026-09-05 13:00:00'],
              dtype='datetime64[us]', freq='h')

Every 5 Hours 30 Minutes

dates=pd.date_range(
    start='2026-09-05 00:00',
    periods=5,
    freq='5h30min'
)

print(dates)

Output

DatetimeIndex(['2026-09-05 00:00:00', '2026-09-05 05:30:00',
               '2026-09-05 11:00:00', '2026-09-05 16:30:00',
               '2026-09-05 22:00:00'],
              dtype='datetime64[us]', freq='330min')

Every Minute

dates=pd.date_range(
    start='2026-09-05 10:00',
    periods=5,
    freq='min'
)

Every Second

dates=pd.date_range(
    start='2026-09-05 10:00',
    periods=5,
    freq='s'
)

Frequency alias update: Use lowercase h for hours and min for minutes. Older examples using H and T are no longer suitable for current Pandas.

Generate Dates Using Today's Date 🔝

Timestamp.today() can supply the current local date and time.

import pandas as pd

today=pd.Timestamp.today().normalize()

dates=pd.date_range(
    end=today,
    periods=10,
    freq='D'
)

print(dates)

This creates exactly ten dates ending with today. Using normalize() changes the current timestamp to midnight before the range is generated.

Next Five Hours from the Current Time

import pandas as pd

now=pd.Timestamp.now()

dates=pd.date_range(
    start=now,
    periods=5,
    freq='h'
)

print(dates)

The actual output depends on the time at which the program is run.

Convert date_range() to a Python List 🔝

date_range() returns a DatetimeIndex. Use tolist() to convert it to a Python list.

import pandas as pd

dates=pd.date_range(
    start='2026-09-01',
    periods=5
)

date_list=dates.tolist()

print(date_list)

The resulting list contains Pandas Timestamp objects.

Convert to Date Strings

date_strings=dates.strftime(
    '%Y-%m-%d'
).tolist()

print(date_strings)

Output

['2026-09-01', '2026-09-02', '2026-09-03',
 '2026-09-04', '2026-09-05']

Include or Exclude Start and End Dates 🔝

The inclusive parameter controls whether the start and end boundaries are included.

The available values are:

'both'
'left'
'right'
'neither'

Include Both Boundaries

dates=pd.date_range(
    start='2026-09-01',
    end='2026-09-05',
    inclusive='both'
)

Output

Exclude the Start Date

dates=pd.date_range(
    start='2026-09-01',
    end='2026-09-05',
    inclusive='right'
)

Output

Exclude the End Date

dates=pd.date_range(
    start='2026-09-01',
    end='2026-09-05',
    inclusive='left'
)

Output

Exclude Both Boundaries

dates=pd.date_range(
    start='2026-09-01',
    end='2026-09-05',
    inclusive='neither'
)

Output

2026-09-02
2026-09-03
2026-09-04

Older Pandas versions used closed='left' or closed='right'. Use inclusive with current Pandas.

Normalize Time to Midnight 🔝

With normalize=True, the start and end values are normalized to midnight before the date range is generated.

import pandas as pd

dates=pd.date_range(
    start='2026-09-05 14:35:20',
    periods=3,
    freq='D',
    normalize=True
)

print(dates)

Output

DatetimeIndex(['2026-09-05', '2026-09-06', '2026-09-07'],
              dtype='datetime64[us]', freq='D')

Without normalization, the time 14:35:20 would be retained in each generated timestamp.

Create a Timezone-Aware Date Range with tz 🔝

The tz parameter creates a timezone-aware DatetimeIndex.

import pandas as pd

dates=pd.date_range(
    start='2026-09-05 09:00',
    periods=3,
    freq='4h',
    tz='Asia/Kolkata'
)

print(dates)

Output

DatetimeIndex(['2026-09-05 09:00:00+05:30',
               '2026-09-05 13:00:00+05:30',
               '2026-09-05 17:00:00+05:30'],
              dtype='datetime64[us, Asia/Kolkata]', freq='4h')

Give the DatetimeIndex a Name 🔝

The name parameter assigns a name to the resulting DatetimeIndex.

dates=pd.date_range(
    start='2026-09-01',
    periods=3,
    name='report_date'
)

print(dates.name)

Output

report_date

The name parameter does not convert the result into a DataFrame. The result remains a DatetimeIndex.

Set Datetime Resolution with unit 🔝

The unit parameter can request a particular datetime resolution. Supported values include seconds, milliseconds, microseconds and nanoseconds.

dates=pd.date_range(
    start='2026-09-05 10:00:00',
    periods=3,
    freq='h',
    unit='s'
)

print(dates.dtype)

Output

datetime64[s]

How Frequency Boundaries Affect the Result 🔝

For anchored frequencies such as month start or month end, the starting value itself may not be a valid timestamp for that frequency.

For example:

dates=pd.date_range(
    start='2026-01-15',
    periods=3,
    freq='MS'
)

print(dates)

Output

DatetimeIndex(['2026-02-01', '2026-03-01', '2026-04-01'],
              dtype='datetime64[us]', freq='MS')

15 January is not a month-start boundary, so the first valid MS timestamp is 1 February.

Similarly, with ME, Pandas moves to the next valid month-end timestamp when necessary.

Current Pandas Frequency Aliases 🔝

Several older frequency aliases have changed. When writing new code, prefer the current forms below.

Current Alias	Meaning	Older Alias to Avoid
`h`	Hour	`H`
`min`	Minute	`T`
`ME`	Month end	`M`
`QE`	Quarter end	`Q`
`YE`	Year end	`Y`
`YE-FEB`	Year ending in February	`A-FEB`

Aliases such as D, B, W, MS, YS, s, ms, us and ns continue to represent their corresponding frequencies.

Common Mistakes with pandas.date_range() 🔝

1. Calling DataFrame.date_range()

Incorrect:

df.date_range(...)

Correct:

pd.date_range(...)

2. Using Old Frequency Aliases

For current Pandas, avoid old code such as:

freq='H'
freq='T'
freq='M'
freq='A-FEB'

Use:

freq='h'
freq='min'
freq='ME'
freq='YE-FEB'

3. Using closed Instead of inclusive

Old:

closed='right'

Current:

inclusive='right'

4. Expecting periods to Mean Days

periods means the number of generated timestamps. Their distance depends on freq.

pd.date_range(
    start='2026-09-01',
    periods=3,
    freq='ME'
)

This generates three month-end timestamps, not three days.

5. Supplying Too Many Conflicting Parameters

The function determines the range from combinations of start, end, periods and freq. Do not try to independently fix all four values when they conflict.

6. Assuming an Anchored Frequency Always Starts on start

For frequencies such as MS or ME, the first result must be a valid boundary for that frequency.

Summary of pandas.date_range() 🔝

pd.date_range() creates a DatetimeIndex.
It is a Pandas function, not a DataFrame method.
Use start and end to define date boundaries.
Use periods to control how many timestamps are generated.
Use freq to control the distance between timestamps.
D generates calendar days and B standard business days.
Use MS for month start and ME for month end.
Use YS for year start and YE for year end.
Use lowercase h for hours and min for minutes.
Multiple frequency units can be combined, such as 5h30min.
inclusive controls whether start and end boundaries are included.
normalize=True resets the start and end times to midnight before generating the range.
tz creates a timezone-aware DatetimeIndex.
name assigns a name to the DatetimeIndex.
unit controls datetime resolution where supported.
tolist() converts the DatetimeIndex to a Python list.
Anchored frequencies such as MS and ME start from the next valid boundary when necessary.
Older aliases such as H, T, M, Y and A-FEB should not be used in current Pandas date_range() examples.

Practice Pandas date and time operations with these exercises.
Exercise 3 on Date and Time
Exercise 3-2 on Date and Time Basics

to_datetime() period_range() DateOffset() strftime()

Pandas Date & Time Pandas Python Tutorials

Subhendu Mohapatra

Author & Instructor at plus2net

I write and maintain practical tutorials on Python, PHP, SQL, JavaScript, HTML, jQuery, and web development at plus2net. The tutorials focus on clear explanations, working examples, and code that readers can test and adapt while learning.

About the author Watch on YouTube

Connect:

Subscribe to our YouTube Channel here

Pandas date_range(): Generate Date and Time Ranges

Syntax of pandas.date_range() 🔝

Basic Date Range Using start and end 🔝

Generate a Fixed Number of Dates with periods 🔝

start + periods

end + periods

start + end + periods

Generate Dates with freq 🔝

Daily and Custom Day Frequencies 🔝

Every Day

Every Three Days

Combine Days and Hours

Generate Business Days 🔝

Month Start and Month End 🔝

Month Start with MS

Month End with ME

Year-End Frequency 🔝

Year Ending in February

Generate Hours, Minutes and Seconds 🔝

Hourly Range

Every 5 Hours 30 Minutes

Every Minute

Every Second

Generate Dates Using Today's Date 🔝

Next Five Hours from the Current Time

Convert date_range() to a Python List 🔝

Convert to Date Strings

Include or Exclude Start and End Dates 🔝

Include Both Boundaries

Exclude the Start Date

Exclude the End Date

Exclude Both Boundaries

Normalize Time to Midnight 🔝

Create a Timezone-Aware Date Range with tz 🔝

Give the DatetimeIndex a Name 🔝

Set Datetime Resolution with unit 🔝

How Frequency Boundaries Affect the Result 🔝

Current Pandas Frequency Aliases 🔝

Common Mistakes with pandas.date_range() 🔝

1. Calling DataFrame.date_range()

2. Using Old Frequency Aliases

3. Using closed Instead of inclusive

4. Expecting periods to Mean Days

5. Supplying Too Many Conflicting Parameters

6. Assuming an Anchored Frequency Always Starts on start

Summary of pandas.date_range() 🔝

Subhendu Mohapatra