fredpy.series class

class fredpy.series(series_id=None, observation_date=None, cache=True)

Creates an instance of fredpy.series that stores information about the specified data series from FRED with the unique series ID code given by series_id.

Parameters

  • series_id (str) – unique FRED series ID. If series_id equals None, an empty fredpy.series instance is created.

  • observation_date (str) – Desired date at which data are observed. Either YYYY-MM-DD or MM-DD-YYYY format. If observation_date is None, today’s date is used.

  • cache (bool) – Whether to store a copy of the downloaded data in memory to avoid repeating calls to the FRED API. Default: True.

Attributes:

data

(Pandas Series) – data values.

date_range

(string) – specifies the dates of the first and last observations.

frequency

(string) – data frequency. ‘Daily’, ‘Weekly’, ‘Monthly’, ‘Quarterly’, or ‘Annual’.

frequency_short

(string) – data frequency. Abbreviated. ‘D’, ‘W’, ‘M’, ‘Q’, ‘SA, or ‘A’.

last_updated

(string) – date series was last updated.

notes

(string) – details about series. Not available for all series.

observation_date

(string) – vintage date at which data are observed.

release

(string) – statistical release containing data.

seasonal_adjustment

(string) – specifies whether the data has been seasonally adjusted.

seasonal_adjustment_short

(string) –specifies whether the data has been seasonally adjusted. Abbreviated.

series_id

(string) – unique FRED series ID code.

source

(string) – original source of the data.

t

(integer) – number corresponding to frequency: 365 for daily, 52 for weekly, 12 for monthly, 4 for quarterly, and 1 for annual.

title

(string) – title of the data series.

units

(string) – units of the data series.

units_short

(string) units of the data series. Abbreviated.

Methods:

apc(log=False, backward=True)

Computes the percentage change in the data over one year.

Parameters

  • log (bool) – If True, computes the percentage change as \(100\cdot\log(x_{t}/x_{t-k})\), where \(k\) is the number of observations per year. If False (default), compute the percentage change as \(100\cdot\left( x_{t}/x_{t-k} - 1\right)\).

  • backward (str) – If True (default), compute percentage change from the previous year. If False, compute percentage change from current to next year.

Returns

fredpy.series

as_frequency(freq=None, method='mean')

Convert a fredpy.series to a lower frequency.

Parameters

  • freq (str) – Abbreviation of desired frequency: ‘D’,’W’,’M’,’Q’,’A’.

  • method (str) – How to resample the data: ‘first’, ‘last’, ‘mean’ (default), ‘median’, ‘min’, ‘max’, ‘sum’

Returns

fredpy.series

bp_filter(low=None, high=None, K=None)

Computes the bandpass (Baxter-King) filter of the data. Returns two fredpy.series instances containing the cyclical and trend components of the data:

new_series_cycle, new_series_trend

Recommendations:

  • Monthly data: low=24, high=84, K=84

  • Quarterly data: low=6, high=32, K=12

  • Annual data: low=1.5, high=8, K=3

Parameters

  • low (int) – Minimum period for oscillations. Default None, recommended value used.

  • high (int) – Maximum period for oscillations. Default None, recommended value used.

  • K (int) – Lead-lag length of the filter. Default None, recommended value used.

Returns

two fredpy.series instances

Note

In computing the bandpass filter, K observations are lost from each end of the original series so the attributes dates, datetimes, and data are 2K elements shorter than their counterparts in the original series.

cf_filter(low=None, high=None)

Computes the Christiano-Fitzgerald filter of the data. Returns two fredpy.series instances containing the cyclical and trend components of the data:

new_series_cycle, new_series_trend

Recommendations:

  • Monthly data: low=18, high=96

  • Quarterly data: low=6, high=32

  • Annual data: low=2, high=8

Parameters

  • low (int) – Minimum period for oscillations. Default None, recommended value used.

  • high (int) – Maximum period for oscillations. Default None, recommended value used.

Returns

two fredpy.series instances

copy()

Returns a copy of the fredpy.series instance.

Parameters

None

Returns

fredpy.series

diff_filter()

Computes the first difference filter of original series. Returns two fredpy.series instances containing the cyclical and trend components of the data:

new_series_cycle, new_series_trend

Parameters

Returns

two fredpy.series instances

Note

In computing the first difference filter, the first observation from the original series is lost so the attributes dates, datetimes, and data are 1 element shorter than their counterparts in the original series.

divide(object2)

Divides the data from the current fredpy series by the data from object2.

Parameters

object2 (fredpy.series) – A fredpy.series instance, number, array, or similar.

Returns

fredpy.series

drop_nan()

Removes NaN values from fredpy series.

Returns

fredpy.series

hp_filter(lamb=None, two_sided=True)

Computes the Hodrick-Prescott filter of the data. Returns two fredpy.series instances containing the cyclical and trend components of the data:

new_series_cycle, new_series_trend

Recommendations:

  • Daily data: lamb= 104976000000

  • Monthly data: lamb=129600

  • Quarterly data: lamb=1600

  • Annual data: lamb=6.25

Parameters

  • lamb (int) – Default None, recommended value used.

  • two_sided (bool) – True (default): Whether to use the two-sided filter or the one-sided version described in Stock and Watson (1999).

Returns

two fredpy.series instances

linear_filter()

Computes a simple linear filter of the data using OLS. Returns two fredpy.series instances containing the cyclical and trend components of the data:

new_series_cycle, new_series_trend

Parameters

Returns

two fredpy.series instances

log()

Computes the natural log of the data.

Parameters

Returns

fredpy.series

ma(length, center=False)

Computes a moving average with window equal to length. If center is True, then the two-sided moving average is computed. Otherwise, the moving average will be one-sided.

Parameters

  • length (int) – window length of the one-sided moving average.

  • center (bool) – False (default): one-sided MA. True: two-sided MA.

Returns

fredpy.series

minus(object2)

Subtracts the data from object2 from the data from the current fredpy series.

Parameters

object2 (fredpy.series) – A fredpy.series instance, number, array, or similar.

Returns

fredpy.series

pc(log=False, backward=True, annualized=False)

Computes the percentage change in the data from the preceding period.

Parameters

  • log (bool) – If True, computes the percentage change as \(100\cdot\log(x_{t}/x_{t-1})\). If False (default), compute the percentage change as \(100\cdot\left( x_{t}/x_{t-1} - 1\right)\).

  • backward (str) – If True, compute percentage change from the previous period. If ‘forward’, compute percentage change from current to subsequent period.

  • annualized (bool) – Default: False: If True, percentage change is computed at an annual rate. E.g., if the data were monthly and log==False, then the annualized percentage change would be: \(100\cdot\left[ \left(x_{t}/x_{t-1}\right)^{12} - 1\right]\).

Returns

fredpy.series

per_capita(total_pop=True)

Transforms the data into per capita terms by dividing by a measure of the total population of the United States.

Parameters

total_pop (str) – If total_pop is True, then use the toal population (Default). Else, use civilian noninstitutional population defined as persons 16 years of age and older.

Returns

fredpy.series

plot(**kwargs)

Equivalent to calling .plot() method on the .data attribute (which is a Pandas Series object). See https://pandas.pydata.org/docs/reference/api/pandas.Series.plot.html for documenation on usage.

plus(object2)

Adds the data from the current fredpy series to the data from object2.

Parameters

object2 (fredpy.series) – A :py:class:fredpy.series instance, number, array, or similar.

Returns

fredpy.series

recent(N)

Restrict the data to the most recent N observations.

Parameters

N (int) – Number of periods to include in the data window.

Returns

fredpy.series

recessions(ax=None, color='0.5', alpha=0.5)

Creates recession bars for plots. Unless ‘ax’ is specified, be used after a plot has been made but before either (1) a new plot is created or (2) a show command is issued.

Parameters

  • ax (matplotlib.axes._subplots.AxesSubplot subplot) – Matplotlib axis object to plot recession bars. Default: None

  • color (str) – Color of the bars. Default: ‘0.5’.

  • alpha (float) – Transparency of the recession bars. Must be between 0 and 1. Default: 0.5.

Returns

times(object2)

Multiplies the data from the current fredpy series with the data from object2.

Parameters

object2 (fredpy.series) – A fredpy.series instance, number, array, or similar.

Returns

fredpy.series

window(win)

Restricts the data to the most recent N observations.

Parameters

win (list) – is an ordered pair: win = [win_min, win_max] where win_min is the date of the minimum date desired and win_max is the date of the maximum date. Date strings must be entered in either YYYY-MM-DD or MM-DD-YYYY format.

Returns

fredpy.series