Time Series / Date functionality

pandas has proven very successful as a tool for working with time series data, especially in the financial data analysis space. With the 0.8 release, we have further improved the time series API in pandas by leaps and bounds. Using the new NumPy datetime64 dtype, we have consolidated a large number of features from other Python libraries like scikits.timeseries as well as created a tremendous amount of new functionality for manipulating time series data.

In working with time series data, we will frequently seek to:

  • generate sequences of fixed-frequency dates and time spans
  • conform or convert time series to a particular frequency
  • compute “relative” dates based on various non-standard time increments (e.g. 5 business days before the last business day of the year), or “roll” dates forward or backward

pandas provides a relatively compact and self-contained set of tools for performing the above tasks.

Create a range of dates:

# 72 hours starting with midnight Jan 1st, 2011
In [1546]: rng = date_range('1/1/2011', periods=72, freq='H')

In [1547]: rng[:5]
Out[1547]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-01 00:00:00, ..., 2011-01-01 04:00:00]
Length: 5, Freq: H, Timezone: None

Index pandas objects with dates:

In [1548]: ts = Series(randn(len(rng)), index=rng)

In [1549]: ts.head()
Out[1549]: 
2011-01-01 00:00:00    0.469112
2011-01-01 01:00:00   -0.282863
2011-01-01 02:00:00   -1.509059
2011-01-01 03:00:00   -1.135632
2011-01-01 04:00:00    1.212112
Freq: H, dtype: float64

Change frequency and fill gaps:

# to 45 minute frequency and forward fill
In [1550]: converted = ts.asfreq('45Min', method='pad')

In [1551]: converted.head()
Out[1551]: 
2011-01-01 00:00:00    0.469112
2011-01-01 00:45:00    0.469112
2011-01-01 01:30:00   -0.282863
2011-01-01 02:15:00   -1.509059
2011-01-01 03:00:00   -1.135632
Freq: 45T, dtype: float64

Resample:

# Daily means
In [1552]: ts.resample('D', how='mean')
Out[1552]: 
2011-01-01   -0.319569
2011-01-02   -0.337703
2011-01-03    0.117258
Freq: D, dtype: float64

Time Stamps vs. Time Spans

Time-stamped data is the most basic type of timeseries data that associates values with points in time. For pandas objects it means using the points in time to create the index

In [1553]: dates = [datetime(2012, 5, 1), datetime(2012, 5, 2), datetime(2012, 5, 3)]

In [1554]: ts = Series(np.random.randn(3), dates)

In [1555]: type(ts.index)
Out[1555]: pandas.tseries.index.DatetimeIndex

In [1556]: ts
Out[1556]: 
2012-05-01   -0.410001
2012-05-02   -0.078638
2012-05-03    0.545952
dtype: float64

However, in many cases it is more natural to associate things like change variables with a time span instead.

For example:

In [1557]: periods = PeriodIndex([Period('2012-01'), Period('2012-02'),
   ......:                        Period('2012-03')])
   ......:

In [1558]: ts = Series(np.random.randn(3), periods)

In [1559]: type(ts.index)
Out[1559]: pandas.tseries.period.PeriodIndex

In [1560]: ts
Out[1560]: 
2012-01   -1.219217
2012-02   -1.226825
2012-03    0.769804
Freq: M, dtype: float64

Starting with 0.8, pandas allows you to capture both representations and convert between them. Under the hood, pandas represents timestamps using instances of Timestamp and sequences of timestamps using instances of DatetimeIndex. For regular time spans, pandas uses Period objects for scalar values and PeriodIndex for sequences of spans. Better support for irregular intervals with arbitrary start and end points are forth-coming in future releases.

Generating Ranges of Timestamps

To generate an index with time stamps, you can use either the DatetimeIndex or Index constructor and pass in a list of datetime objects:

In [1561]: dates = [datetime(2012, 5, 1), datetime(2012, 5, 2), datetime(2012, 5, 3)]

In [1562]: index = DatetimeIndex(dates)

In [1563]: index # Note the frequency information
Out[1563]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2012-05-01 00:00:00, ..., 2012-05-03 00:00:00]
Length: 3, Freq: None, Timezone: None

In [1564]: index = Index(dates)

In [1565]: index # Automatically converted to DatetimeIndex
Out[1565]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2012-05-01 00:00:00, ..., 2012-05-03 00:00:00]
Length: 3, Freq: None, Timezone: None

Practically, this becomes very cumbersome because we often need a very long index with a large number of timestamps. If we need timestamps on a regular frequency, we can use the pandas functions date_range and bdate_range to create timestamp indexes.

In [1566]: index = date_range('2000-1-1', periods=1000, freq='M')

In [1567]: index
Out[1567]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2000-01-31 00:00:00, ..., 2083-04-30 00:00:00]
Length: 1000, Freq: M, Timezone: None

In [1568]: index = bdate_range('2012-1-1', periods=250)

In [1569]: index
Out[1569]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2012-01-02 00:00:00, ..., 2012-12-14 00:00:00]
Length: 250, Freq: B, Timezone: None

Convenience functions like date_range and bdate_range utilize a variety of frequency aliases. The default frequency for date_range is a calendar day while the default for bdate_range is a business day

In [1570]: start = datetime(2011, 1, 1)

In [1571]: end = datetime(2012, 1, 1)

In [1572]: rng = date_range(start, end)

In [1573]: rng
Out[1573]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-01 00:00:00, ..., 2012-01-01 00:00:00]
Length: 366, Freq: D, Timezone: None

In [1574]: rng = bdate_range(start, end)

In [1575]: rng
Out[1575]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-03 00:00:00, ..., 2011-12-30 00:00:00]
Length: 260, Freq: B, Timezone: None

date_range and bdate_range makes it easy to generate a range of dates using various combinations of parameters like start, end, periods, and freq:

In [1576]: date_range(start, end, freq='BM')
Out[1576]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-31 00:00:00, ..., 2011-12-30 00:00:00]
Length: 12, Freq: BM, Timezone: None

In [1577]: date_range(start, end, freq='W')
Out[1577]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-02 00:00:00, ..., 2012-01-01 00:00:00]
Length: 53, Freq: W-SUN, Timezone: None

In [1578]: bdate_range(end=end, periods=20)
Out[1578]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-12-05 00:00:00, ..., 2011-12-30 00:00:00]
Length: 20, Freq: B, Timezone: None

In [1579]: bdate_range(start=start, periods=20)
Out[1579]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-03 00:00:00, ..., 2011-01-28 00:00:00]
Length: 20, Freq: B, Timezone: None

The start and end dates are strictly inclusive. So it will not generate any dates outside of those dates if specified.

DatetimeIndex

One of the main uses for DatetimeIndex is as an index for pandas objects. The DatetimeIndex class contains many timeseries related optimizations:

  • A large range of dates for various offsets are pre-computed and cached under the hood in order to make generating subsequent date ranges very fast (just have to grab a slice)
  • Fast shifting using the shift and tshift method on pandas objects
  • Unioning of overlapping DatetimeIndex objects with the same frequency is very fast (important for fast data alignment)
  • Quick access to date fields via properties such as year, month, etc.
  • Regularization functions like snap and very fast asof logic

DatetimeIndex can be used like a regular index and offers all of its intelligent functionality like selection, slicing, etc.

In [1580]: rng = date_range(start, end, freq='BM')

In [1581]: ts = Series(randn(len(rng)), index=rng)

In [1582]: ts.index
Out[1582]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-31 00:00:00, ..., 2011-12-30 00:00:00]
Length: 12, Freq: BM, Timezone: None

In [1583]: ts[:5].index
Out[1583]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-31 00:00:00, ..., 2011-05-31 00:00:00]
Length: 5, Freq: BM, Timezone: None

In [1584]: ts[::2].index
Out[1584]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-31 00:00:00, ..., 2011-11-30 00:00:00]
Length: 6, Freq: 2BM, Timezone: None

You can pass in dates and strings that parses to dates as indexing parameters:

In [1585]: ts['1/31/2011']
Out[1585]: -1.2812473076599531

In [1586]: ts[datetime(2011, 12, 25):]
Out[1586]: 
2011-12-30    0.687738
Freq: BM, dtype: float64

In [1587]: ts['10/31/2011':'12/31/2011']
Out[1587]: 
2011-10-31    0.149748
2011-11-30   -0.732339
2011-12-30    0.687738
Freq: BM, dtype: float64

A truncate convenience function is provided that is equivalent to slicing:

In [1588]: ts.truncate(before='10/31/2011', after='12/31/2011')
Out[1588]: 
2011-10-31    0.149748
2011-11-30   -0.732339
2011-12-30    0.687738
Freq: BM, dtype: float64

To provide convenience for accessing longer time series, you can also pass in the year or year and month as strings:

In [1589]: ts['2011']
Out[1589]: 
2011-01-31   -1.281247
2011-02-28   -0.727707
2011-03-31   -0.121306
2011-04-29   -0.097883
2011-05-31    0.695775
2011-06-30    0.341734
2011-07-29    0.959726
2011-08-31   -1.110336
2011-09-30   -0.619976
2011-10-31    0.149748
2011-11-30   -0.732339
2011-12-30    0.687738
Freq: BM, dtype: float64

In [1590]: ts['2011-6']
Out[1590]: 
2011-06-30    0.341734
Freq: BM, dtype: float64

Even complicated fancy indexing that breaks the DatetimeIndex’s frequency regularity will result in a DatetimeIndex (but frequency is lost):

In [1591]: ts[[0, 2, 6]].index
Out[1591]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-31 00:00:00, ..., 2011-07-29 00:00:00]
Length: 3, Freq: None, Timezone: None

DatetimeIndex objects has all the basic functionality of regular Index objects and a smorgasbord of advanced timeseries-specific methods for easy frequency processing.

Note

While pandas does not force you to have a sorted date index, some of these methods may have unexpected or incorrect behavior if the dates are unsorted. So please be careful.

DateOffset objects

In the preceding examples, we created DatetimeIndex objects at various frequencies by passing in frequency strings like ‘M’, ‘W’, and ‘BM to the freq keyword. Under the hood, these frequency strings are being translated into an instance of pandas DateOffset, which represents a regular frequency increment. Specific offset logic like “month”, “business day”, or “one hour” is represented in its various subclasses.

Class name Description
DateOffset Generic offset class, defaults to 1 calendar day
BDay business day (weekday)
Week one week, optionally anchored on a day of the week
WeekOfMonth the x-th day of the y-th week of each month
MonthEnd calendar month end
MonthBegin calendar month begin
BMonthEnd business month end
BMonthBegin business month begin
QuarterEnd calendar quarter end
QuarterBegin calendar quarter begin
BQuarterEnd business quarter end
BQuarterBegin business quarter begin
YearEnd calendar year end
YearBegin calendar year begin
BYearEnd business year end
BYearBegin business year begin
Hour one hour
Minute one minute
Second one second
Milli one millisecond
Micro one microsecond

The basic DateOffset takes the same arguments as dateutil.relativedelta, which works like:

In [1592]: d = datetime(2008, 8, 18)

In [1593]: d + relativedelta(months=4, days=5)
Out[1593]: datetime.datetime(2008, 12, 23, 0, 0)

We could have done the same thing with DateOffset:

In [1594]: from pandas.tseries.offsets import *

In [1595]: d + DateOffset(months=4, days=5)
Out[1595]: datetime.datetime(2008, 12, 23, 0, 0)

The key features of a DateOffset object are:

  • it can be added / subtracted to/from a datetime object to obtain a shifted date
  • it can be multiplied by an integer (positive or negative) so that the increment will be applied multiple times
  • it has rollforward and rollback methods for moving a date forward or backward to the next or previous “offset date”

Subclasses of DateOffset define the apply function which dictates custom date increment logic, such as adding business days:

class BDay(DateOffset):
    """DateOffset increments between business days"""
    def apply(self, other):
        ...
In [1596]: d - 5 * BDay()
Out[1596]: datetime.datetime(2008, 8, 11, 0, 0)

In [1597]: d + BMonthEnd()
Out[1597]: datetime.datetime(2008, 8, 29, 0, 0)

The rollforward and rollback methods do exactly what you would expect:

In [1598]: d
Out[1598]: datetime.datetime(2008, 8, 18, 0, 0)

In [1599]: offset = BMonthEnd()

In [1600]: offset.rollforward(d)
Out[1600]: datetime.datetime(2008, 8, 29, 0, 0)

In [1601]: offset.rollback(d)
Out[1601]: datetime.datetime(2008, 7, 31, 0, 0)

It’s definitely worth exploring the pandas.tseries.offsets module and the various docstrings for the classes.

Parametric offsets

Some of the offsets can be “parameterized” when created to result in different behavior. For example, the Week offset for generating weekly data accepts a weekday parameter which results in the generated dates always lying on a particular day of the week:

In [1602]: d + Week()
Out[1602]: datetime.datetime(2008, 8, 25, 0, 0)

In [1603]: d + Week(weekday=4)
Out[1603]: datetime.datetime(2008, 8, 22, 0, 0)

In [1604]: (d + Week(weekday=4)).weekday()
Out[1604]: 4

Another example is parameterizing YearEnd with the specific ending month:

In [1605]: d + YearEnd()
Out[1605]: datetime.datetime(2008, 12, 31, 0, 0)

In [1606]: d + YearEnd(month=6)
Out[1606]: datetime.datetime(2009, 6, 30, 0, 0)

Offset Aliases

A number of string aliases are given to useful common time series frequencies. We will refer to these aliases as offset aliases (referred to as time rules prior to v0.8.0).

Alias Description
B business day frequency
D calendar day frequency
W weekly frequency
M month end frequency
BM business month end frequency
MS month start frequency
BMS business month start frequency
Q quarter end frequency
BQ business quarter endfrequency
QS quarter start frequency
BQS business quarter start frequency
A year end frequency
BA business year end frequency
AS year start frequency
BAS business year start frequency
H hourly frequency
T minutely frequency
S secondly frequency
L milliseonds
U microseconds

Combining Aliases

As we have seen previously, the alias and the offset instance are fungible in most functions:

In [1607]: date_range(start, periods=5, freq='B')
Out[1607]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-03 00:00:00, ..., 2011-01-07 00:00:00]
Length: 5, Freq: B, Timezone: None

In [1608]: date_range(start, periods=5, freq=BDay())
Out[1608]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-03 00:00:00, ..., 2011-01-07 00:00:00]
Length: 5, Freq: B, Timezone: None

You can combine together day and intraday offsets:

In [1609]: date_range(start, periods=10, freq='2h20min')
Out[1609]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-01 00:00:00, ..., 2011-01-01 21:00:00]
Length: 10, Freq: 140T, Timezone: None

In [1610]: date_range(start, periods=10, freq='1D10U')
Out[1610]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2011-01-01 00:00:00, ..., 2011-01-10 00:00:00.000090]
Length: 10, Freq: 86400000010U, Timezone: None

Anchored Offsets

For some frequencies you can specify an anchoring suffix:

Alias Description
W-SUN weekly frequency (sundays). Same as ‘W’
W-MON weekly frequency (mondays)
W-TUE weekly frequency (tuesdays)
W-WED weekly frequency (wednesdays)
W-THU weekly frequency (thursdays)
W-FRI weekly frequency (fridays)
W-SAT weekly frequency (saturdays)
(B)Q(S)-DEC quarterly frequency, year ends in December. Same as ‘Q’
(B)Q(S)-JAN quarterly frequency, year ends in January
(B)Q(S)-FEB quarterly frequency, year ends in February
(B)Q(S)-MAR quarterly frequency, year ends in March
(B)Q(S)-APR quarterly frequency, year ends in April
(B)Q(S)-MAY quarterly frequency, year ends in May
(B)Q(S)-JUN quarterly frequency, year ends in June
(B)Q(S)-JUL quarterly frequency, year ends in July
(B)Q(S)-AUG quarterly frequency, year ends in August
(B)Q(S)-SEP quarterly frequency, year ends in September
(B)Q(S)-OCT quarterly frequency, year ends in October
(B)Q(S)-NOV quarterly frequency, year ends in November
(B)A(S)-DEC annual frequency, anchored end of December. Same as ‘A’
(B)A(S)-JAN annual frequency, anchored end of January
(B)A(S)-FEB annual frequency, anchored end of February
(B)A(S)-MAR annual frequency, anchored end of March
(B)A(S)-APR annual frequency, anchored end of April
(B)A(S)-MAY annual frequency, anchored end of May
(B)A(S)-JUN annual frequency, anchored end of June
(B)A(S)-JUL annual frequency, anchored end of July
(B)A(S)-AUG annual frequency, anchored end of August
(B)A(S)-SEP annual frequency, anchored end of September
(B)A(S)-OCT annual frequency, anchored end of October
(B)A(S)-NOV annual frequency, anchored end of November

These can be used as arguments to date_range, bdate_range, constructors for DatetimeIndex, as well as various other timeseries-related functions in pandas.

Legacy Aliases

Note that prior to v0.8.0, time rules had a slightly different look. Pandas will continue to support the legacy time rules for the time being but it is strongly recommended that you switch to using the new offset aliases.

Legacy Time Rule Offset Alias
WEEKDAY B
EOM BM
W@MON W-MON
W@TUE W-TUE
W@WED W-WED
W@THU W-THU
W@FRI W-FRI
W@SAT W-SAT
W@SUN W-SUN
Q@JAN BQ-JAN
Q@FEB BQ-FEB
Q@MAR BQ-MAR
A@JAN BA-JAN
A@FEB BA-FEB
A@MAR BA-MAR
A@APR BA-APR
A@MAY BA-MAY
A@JUN BA-JUN
A@JUL BA-JUL
A@AUG BA-AUG
A@SEP BA-SEP
A@OCT BA-OCT
A@NOV BA-NOV
A@DEC BA-DEC
min T
ms L
us U

As you can see, legacy quarterly and annual frequencies are business quarter and business year ends. Please also note the legacy time rule for milliseconds ms versus the new offset alias for month start MS. This means that offset alias parsing is case sensitive.

Up- and downsampling

With 0.8, pandas introduces simple, powerful, and efficient functionality for performing resampling operations during frequency conversion (e.g., converting secondly data into 5-minutely data). This is extremely common in, but not limited to, financial applications.

See some cookbook examples for some advanced strategies

In [1621]: rng = date_range('1/1/2012', periods=100, freq='S')

In [1622]: ts = Series(randint(0, 500, len(rng)), index=rng)

In [1623]: ts.resample('5Min', how='sum')
Out[1623]: 
2012-01-01    25792
Freq: 5T, dtype: int64

The resample function is very flexible and allows you to specify many different parameters to control the frequency conversion and resampling operation.

The how parameter can be a function name or numpy array function that takes an array and produces aggregated values:

In [1624]: ts.resample('5Min') # default is mean
Out[1624]: 
2012-01-01    257.92
Freq: 5T, dtype: float64

In [1625]: ts.resample('5Min', how='ohlc')
Out[1625]: 
            open  high  low  close
2012-01-01   230   492    0    214

In [1626]: ts.resample('5Min', how=np.max)
Out[1626]: 
2012-01-01   NaN
Freq: 5T, dtype: float64

Any function available via dispatching can be given to the how parameter by name, including sum, mean, std, max, min, median, first, last, ohlc.

For downsampling, closed can be set to ‘left’ or ‘right’ to specify which end of the interval is closed:

In [1627]: ts.resample('5Min', closed='right')
Out[1627]: 
2011-12-31 23:55:00    230.00000
2012-01-01 00:00:00    258.20202
Freq: 5T, dtype: float64

In [1628]: ts.resample('5Min', closed='left')
Out[1628]: 
2012-01-01    257.92
Freq: 5T, dtype: float64

For upsampling, the fill_method and limit parameters can be specified to interpolate over the gaps that are created:

# from secondly to every 250 milliseconds
In [1629]: ts[:2].resample('250L')
Out[1629]: 
2012-01-01 00:00:00           230
2012-01-01 00:00:00.250000    NaN
2012-01-01 00:00:00.500000    NaN
2012-01-01 00:00:00.750000    NaN
2012-01-01 00:00:01           202
Freq: 250L, dtype: float64

In [1630]: ts[:2].resample('250L', fill_method='pad')
Out[1630]: 
2012-01-01 00:00:00           230
2012-01-01 00:00:00.250000    230
2012-01-01 00:00:00.500000    230
2012-01-01 00:00:00.750000    230
2012-01-01 00:00:01           202
Freq: 250L, dtype: int64

In [1631]: ts[:2].resample('250L', fill_method='pad', limit=2)
Out[1631]: 
2012-01-01 00:00:00           230
2012-01-01 00:00:00.250000    230
2012-01-01 00:00:00.500000    230
2012-01-01 00:00:00.750000    NaN
2012-01-01 00:00:01           202
Freq: 250L, dtype: float64

Parameters like label and loffset are used to manipulate the resulting labels. label specifies whether the result is labeled with the beginning or the end of the interval. loffset performs a time adjustment on the output labels.

In [1632]: ts.resample('5Min') # by default label='right'
Out[1632]: 
2012-01-01    257.92
Freq: 5T, dtype: float64

In [1633]: ts.resample('5Min', label='left')
Out[1633]: 
2012-01-01    257.92
Freq: 5T, dtype: float64

In [1634]: ts.resample('5Min', label='left', loffset='1s')
Out[1634]: 
2012-01-01 00:00:01    257.92
dtype: float64

The axis parameter can be set to 0 or 1 and allows you to resample the specified axis for a DataFrame.

kind can be set to ‘timestamp’ or ‘period’ to convert the resulting index to/from time-stamp and time-span representations. By default resample retains the input representation.

convention can be set to ‘start’ or ‘end’ when resampling period data (detail below). It specifies how low frequency periods are converted to higher frequency periods.

Note that 0.8 marks a watershed in the timeseries functionality in pandas. In previous versions, resampling had to be done using a combination of date_range, groupby with asof, and then calling an aggregation function on the grouped object. This was not nearly convenient or performant as the new pandas timeseries API.

Time Span Representation

Regular intervals of time are represented by Period objects in pandas while sequences of Period objects are collected in a PeriodIndex, which can be created with the convenience function period_range.

Period

A Period represents a span of time (e.g., a day, a month, a quarter, etc). It can be created using a frequency alias:

In [1635]: Period('2012', freq='A-DEC')
Out[1635]: Period('2012', 'A-DEC')

In [1636]: Period('2012-1-1', freq='D')
Out[1636]: Period('2012-01-01', 'D')

In [1637]: Period('2012-1-1 19:00', freq='H')
Out[1637]: Period('2012-01-01 19:00', 'H')

Unlike time stamped data, pandas does not support frequencies at multiples of DateOffsets (e.g., ‘3Min’) for periods.

Adding and subtracting integers from periods shifts the period by its own frequency.

In [1638]: p = Period('2012', freq='A-DEC')

In [1639]: p + 1
Out[1639]: Period('2013', 'A-DEC')

In [1640]: p - 3
Out[1640]: Period('2009', 'A-DEC')

Taking the difference of Period instances with the same frequency will return the number of frequency units between them:

In [1641]: Period('2012', freq='A-DEC') - Period('2002', freq='A-DEC')
Out[1641]: 10

PeriodIndex and period_range

Regular sequences of Period objects can be collected in a PeriodIndex, which can be constructed using the period_range convenience function:

In [1642]: prng = period_range('1/1/2011', '1/1/2012', freq='M')

In [1643]: prng
Out[1643]: 
<class 'pandas.tseries.period.PeriodIndex'>
freq: M
[2011-01, ..., 2012-01]
length: 13

The PeriodIndex constructor can also be used directly:

In [1644]: PeriodIndex(['2011-1', '2011-2', '2011-3'], freq='M')
Out[1644]: 
<class 'pandas.tseries.period.PeriodIndex'>
freq: M
[2011-01, ..., 2011-03]
length: 3

Just like DatetimeIndex, a PeriodIndex can also be used to index pandas objects:

In [1645]: Series(randn(len(prng)), prng)
Out[1645]: 
2011-01    0.301624
2011-02   -1.460489
2011-03    0.610679
2011-04    1.195856
2011-05   -0.008820
2011-06   -0.045729
2011-07   -1.051015
2011-08   -0.422924
2011-09   -0.028361
2011-10   -0.782386
2011-11    0.861980
2011-12    1.438604
2012-01   -0.525492
Freq: M, dtype: float64

Frequency Conversion and Resampling with PeriodIndex

The frequency of Periods and PeriodIndex can be converted via the asfreq method. Let’s start with the fiscal year 2011, ending in December:

In [1646]: p = Period('2011', freq='A-DEC')

In [1647]: p
Out[1647]: Period('2011', 'A-DEC')

We can convert it to a monthly frequency. Using the how parameter, we can specify whether to return the starting or ending month:

In [1648]: p.asfreq('M', how='start')
Out[1648]: Period('2011-01', 'M')

In [1649]: p.asfreq('M', how='end')
Out[1649]: Period('2011-12', 'M')

The shorthands ‘s’ and ‘e’ are provided for convenience:

In [1650]: p.asfreq('M', 's')
Out[1650]: Period('2011-01', 'M')

In [1651]: p.asfreq('M', 'e')
Out[1651]: Period('2011-12', 'M')

Converting to a “super-period” (e.g., annual frequency is a super-period of quarterly frequency) automatically returns the super-period that includes the input period:

In [1652]: p = Period('2011-12', freq='M')

In [1653]: p.asfreq('A-NOV')
Out[1653]: Period('2012', 'A-NOV')

Note that since we converted to an annual frequency that ends the year in November, the monthly period of December 2011 is actually in the 2012 A-NOV period.

Period conversions with anchored frequencies are particularly useful for working with various quarterly data common to economics, business, and other fields. Many organizations define quarters relative to the month in which their fiscal year start and ends. Thus, first quarter of 2011 could start in 2010 or a few months into 2011. Via anchored frequencies, pandas works all quarterly frequencies Q-JAN through Q-DEC.

Q-DEC define regular calendar quarters:

In [1654]: p = Period('2012Q1', freq='Q-DEC')

In [1655]: p.asfreq('D', 's')
Out[1655]: Period('2012-01-01', 'D')

In [1656]: p.asfreq('D', 'e')
Out[1656]: Period('2012-03-31', 'D')

Q-MAR defines fiscal year end in March:

In [1657]: p = Period('2011Q4', freq='Q-MAR')

In [1658]: p.asfreq('D', 's')
Out[1658]: Period('2011-01-01', 'D')

In [1659]: p.asfreq('D', 'e')
Out[1659]: Period('2011-03-31', 'D')

Converting between Representations

Timestamped data can be converted to PeriodIndex-ed data using to_period and vice-versa using to_timestamp:

In [1660]: rng = date_range('1/1/2012', periods=5, freq='M')

In [1661]: ts = Series(randn(len(rng)), index=rng)

In [1662]: ts
Out[1662]: 
2012-01-31   -1.684469
2012-02-29    0.550605
2012-03-31    0.091955
2012-04-30    0.891713
2012-05-31    0.807078
Freq: M, dtype: float64

In [1663]: ps = ts.to_period()

In [1664]: ps
Out[1664]: 
2012-01   -1.684469
2012-02    0.550605
2012-03    0.091955
2012-04    0.891713
2012-05    0.807078
Freq: M, dtype: float64

In [1665]: ps.to_timestamp()
Out[1665]: 
2012-01-01   -1.684469
2012-02-01    0.550605
2012-03-01    0.091955
2012-04-01    0.891713
2012-05-01    0.807078
Freq: MS, dtype: float64

Remember that ‘s’ and ‘e’ can be used to return the timestamps at the start or end of the period:

In [1666]: ps.to_timestamp('D', how='s')
Out[1666]: 
2012-01-01   -1.684469
2012-02-01    0.550605
2012-03-01    0.091955
2012-04-01    0.891713
2012-05-01    0.807078
Freq: MS, dtype: float64

Converting between period and timestamp enables some convenient arithmetic functions to be used. In the following example, we convert a quarterly frequency with year ending in November to 9am of the end of the month following the quarter end:

In [1667]: prng = period_range('1990Q1', '2000Q4', freq='Q-NOV')

In [1668]: ts = Series(randn(len(prng)), prng)

In [1669]: ts.index = (prng.asfreq('M', 'e') + 1).asfreq('H', 's') + 9

In [1670]: ts.head()
Out[1670]: 
1990-03-01 09:00    0.221441
1990-06-01 09:00   -0.113139
1990-09-01 09:00   -1.812900
1990-12-01 09:00   -0.053708
1991-03-01 09:00   -0.114574
Freq: H, dtype: float64

Time Zone Handling

Using pytz, pandas provides rich support for working with timestamps in different time zones. By default, pandas objects are time zone unaware:

In [1671]: rng = date_range('3/6/2012 00:00', periods=15, freq='D')

In [1672]: print(rng.tz)
None

To supply the time zone, you can use the tz keyword to date_range and other functions:

In [1673]: rng_utc = date_range('3/6/2012 00:00', periods=10, freq='D', tz='UTC')

In [1674]: print(rng_utc.tz)
UTC

Timestamps, like Python’s datetime.datetime object can be either time zone naive or time zone aware. Naive time series and DatetimeIndex objects can be localized using tz_localize:

In [1675]: ts = Series(randn(len(rng)), rng)

In [1676]: ts_utc = ts.tz_localize('UTC')

In [1677]: ts_utc
Out[1677]: 
2012-03-06 00:00:00+00:00   -0.114722
2012-03-07 00:00:00+00:00    0.168904
2012-03-08 00:00:00+00:00   -0.048048
2012-03-09 00:00:00+00:00    0.801196
2012-03-10 00:00:00+00:00    1.392071
2012-03-11 00:00:00+00:00   -0.048788
2012-03-12 00:00:00+00:00   -0.808838
2012-03-13 00:00:00+00:00   -1.003677
2012-03-14 00:00:00+00:00   -0.160766
2012-03-15 00:00:00+00:00    1.758853
2012-03-16 00:00:00+00:00    0.729195
2012-03-17 00:00:00+00:00    1.359732
2012-03-18 00:00:00+00:00    2.006296
2012-03-19 00:00:00+00:00    0.870210
2012-03-20 00:00:00+00:00    0.043464
Freq: D, dtype: float64

You can use the tz_convert method to convert pandas objects to convert tz-aware data to another time zone:

In [1678]: ts_utc.tz_convert('US/Eastern')
Out[1678]: 
2012-03-05 19:00:00-05:00   -0.114722
2012-03-06 19:00:00-05:00    0.168904
2012-03-07 19:00:00-05:00   -0.048048
2012-03-08 19:00:00-05:00    0.801196
2012-03-09 19:00:00-05:00    1.392071
2012-03-10 19:00:00-05:00   -0.048788
2012-03-11 20:00:00-04:00   -0.808838
2012-03-12 20:00:00-04:00   -1.003677
2012-03-13 20:00:00-04:00   -0.160766
2012-03-14 20:00:00-04:00    1.758853
2012-03-15 20:00:00-04:00    0.729195
2012-03-16 20:00:00-04:00    1.359732
2012-03-17 20:00:00-04:00    2.006296
2012-03-18 20:00:00-04:00    0.870210
2012-03-19 20:00:00-04:00    0.043464
Freq: D, dtype: float64

Under the hood, all timestamps are stored in UTC. Scalar values from a DatetimeIndex with a time zone will have their fields (day, hour, minute) localized to the time zone. However, timestamps with the same UTC value are still considered to be equal even if they are in different time zones:

In [1679]: rng_eastern = rng_utc.tz_convert('US/Eastern')

In [1680]: rng_berlin = rng_utc.tz_convert('Europe/Berlin')

In [1681]: rng_eastern[5]
Out[1681]: <Timestamp: 2012-03-10 19:00:00-0500 EST, tz=US/Eastern>

In [1682]: rng_berlin[5]
Out[1682]: <Timestamp: 2012-03-11 01:00:00+0100 CET, tz=Europe/Berlin>

In [1683]: rng_eastern[5] == rng_berlin[5]
Out[1683]: True

Like Series, DataFrame, and DatetimeIndex, Timestamps can be converted to other time zones using tz_convert:

In [1684]: rng_eastern[5]
Out[1684]: <Timestamp: 2012-03-10 19:00:00-0500 EST, tz=US/Eastern>

In [1685]: rng_berlin[5]
Out[1685]: <Timestamp: 2012-03-11 01:00:00+0100 CET, tz=Europe/Berlin>

In [1686]: rng_eastern[5].tz_convert('Europe/Berlin')
Out[1686]: <Timestamp: 2012-03-11 01:00:00+0100 CET, tz=Europe/Berlin>

Localization of Timestamps functions just like DatetimeIndex and TimeSeries:

In [1687]: rng[5]
Out[1687]: <Timestamp: 2012-03-11 00:00:00>

In [1688]: rng[5].tz_localize('Asia/Shanghai')
Out[1688]: <Timestamp: 2012-03-11 00:00:00+0800 CST, tz=Asia/Shanghai>

Operations between TimeSeries in difficult time zones will yield UTC TimeSeries, aligning the data on the UTC timestamps:

In [1689]: eastern = ts_utc.tz_convert('US/Eastern')

In [1690]: berlin = ts_utc.tz_convert('Europe/Berlin')

In [1691]: result = eastern + berlin

In [1692]: result
Out[1692]: 
2012-03-06 00:00:00+00:00   -0.229443
2012-03-07 00:00:00+00:00    0.337809
2012-03-08 00:00:00+00:00   -0.096096
2012-03-09 00:00:00+00:00    1.602392
2012-03-10 00:00:00+00:00    2.784142
2012-03-11 00:00:00+00:00   -0.097575
2012-03-12 00:00:00+00:00   -1.617677
2012-03-13 00:00:00+00:00   -2.007353
2012-03-14 00:00:00+00:00   -0.321532
2012-03-15 00:00:00+00:00    3.517706
2012-03-16 00:00:00+00:00    1.458389
2012-03-17 00:00:00+00:00    2.719465
2012-03-18 00:00:00+00:00    4.012592
2012-03-19 00:00:00+00:00    1.740419
2012-03-20 00:00:00+00:00    0.086928
Freq: D, dtype: float64

In [1693]: result.index
Out[1693]: 
<class 'pandas.tseries.index.DatetimeIndex'>
[2012-03-06 00:00:00, ..., 2012-03-20 00:00:00]
Length: 15, Freq: D, Timezone: UTC

Time Deltas

Timedeltas are differences in times, expressed in difference units, e.g. days,hours,minutes,seconds. They can be both positive and negative.

In [1694]: from datetime import datetime, timedelta

In [1695]: s  = Series(date_range('2012-1-1', periods=3, freq='D'))

In [1696]: td = Series([ timedelta(days=i) for i in range(3) ])

In [1697]: df = DataFrame(dict(A = s, B = td))

In [1698]: df
Out[1698]: 
                    A                B
0 2012-01-01 00:00:00         00:00:00
1 2012-01-02 00:00:00 1 days, 00:00:00
2 2012-01-03 00:00:00 2 days, 00:00:00

In [1699]: df['C'] = df['A'] + df['B']

In [1700]: df
Out[1700]: 
                    A                B                   C
0 2012-01-01 00:00:00         00:00:00 2012-01-01 00:00:00
1 2012-01-02 00:00:00 1 days, 00:00:00 2012-01-03 00:00:00
2 2012-01-03 00:00:00 2 days, 00:00:00 2012-01-05 00:00:00

In [1701]: df.dtypes
Out[1701]: 
A     datetime64[ns]
B    timedelta64[ns]
C     datetime64[ns]
dtype: object

In [1702]: s - s.max()
Out[1702]: 
0   -2 days, 00:00:00
1   -1 days, 00:00:00
2            00:00:00
dtype: timedelta64[ns]

In [1703]: s - datetime(2011,1,1,3,5)
Out[1703]: 
0   364 days, 20:55:00
1   365 days, 20:55:00
2   366 days, 20:55:00
dtype: timedelta64[ns]

In [1704]: s + timedelta(minutes=5)
Out[1704]: 
0   2012-01-01 00:05:00
1   2012-01-02 00:05:00
2   2012-01-03 00:05:00
dtype: datetime64[ns]

Series of timedeltas with NaT values are supported

In [1705]: y = s - s.shift()

In [1706]: y
Out[1706]: 
0                NaT
1   1 days, 00:00:00
2   1 days, 00:00:00
dtype: timedelta64[ns]

The can be set to NaT using np.nan analagously to datetimes

In [1707]: y[1] = np.nan

In [1708]: y
Out[1708]: 
0                NaT
1                NaT
2   1 days, 00:00:00
dtype: timedelta64[ns]
WARNING: Output cache limit (currently 1000 entries) hit.
Flushing cache and resetting history counter...
The only history variables available will be _,__,___ and _1
with the current result.

Operands can also appear in a reversed order (a singluar object operated with a Series)

In [1709]: s.max() - s
Out[1709]: 
0   2 days, 00:00:00
1   1 days, 00:00:00
2           00:00:00
dtype: timedelta64[ns]

In [1710]: datetime(2011,1,1,3,5) - s
Out[1710]: 
0   -364 days, 20:55:00
1   -365 days, 20:55:00
2   -366 days, 20:55:00
dtype: timedelta64[ns]

In [1711]: timedelta(minutes=5) + s
Out[1711]: 
0   2012-01-01 00:05:00
1   2012-01-02 00:05:00
2   2012-01-03 00:05:00
dtype: datetime64[ns]

Some timedelta numeric like operations are supported.

In [1712]: td - timedelta(minutes=5,seconds=5,microseconds=5)
Out[1712]: 
0          -00:05:05.000005
1           23:54:54.999995
2   1 days, 23:54:54.999995
dtype: timedelta64[ns]

min, max and the corresponding idxmin, idxmax operations are support on frames

In [1713]: df = DataFrame(dict(A = s - Timestamp('20120101')-timedelta(minutes=5,seconds=5),
   ......:                     B = s - Series(date_range('2012-1-2', periods=3, freq='D'))))
   ......:

In [1714]: df
Out[1714]: 
                 A                 B
0        -00:05:05 -1 days, 00:00:00
1         23:54:55 -1 days, 00:00:00
2 1 days, 23:54:55 -1 days, 00:00:00

In [1715]: df.min()
Out[1715]: 
A           -00:05:05
B   -1 days, 00:00:00
dtype: timedelta64[ns]

In [1716]: df.min(axis=1)
Out[1716]: 
0   -1 days, 00:00:00
1   -1 days, 00:00:00
2   -1 days, 00:00:00
dtype: timedelta64[ns]

In [1717]: df.idxmin()
Out[1717]: 
A    0
B    0
dtype: int64

In [1718]: df.idxmax()
Out[1718]: 
A    2
B    0
dtype: int64

min, max operations are support on series, these return a single element timedelta64[ns] Series (this avoids having to deal with numpy timedelta64 issues). idxmin, idxmax are supported as well.

In [1719]: df.min().max()
Out[1719]: 
0   -00:05:05
dtype: timedelta64[ns]

In [1720]: df.min(axis=1).min()
Out[1720]: 
0   -1 days, 00:00:00
dtype: timedelta64[ns]

In [1721]: df.min().idxmax()
Out[1721]: 'A'

In [1722]: df.min(axis=1).idxmin()
Out[1722]: 0