Introduction

This vignette is intended to introduce the user to fredr functions for the Series endpoint of the FRED API.

FRED series are the primary data objects in the FRED database. Each FRED series is assigned a string identifier. For example:

Civilian Unemployment Rate (series_id = "UNRATE")
Real Gross National Product (series_id = "GNPCA")
Effective Federal Funds Rate (series_id = "FEDFUNDS")

FRED series are assigned categories and tags attributes for organization and classification. The following examples illustrate usage of the Series endpoint functions in fredr.

Retrieve observations

The function fredr(), an alias for fredr_series_observations(), is the core function in fredr for fetching FRED data series. See the Get started article and ?fredr() for many usage examples.

fredr(
  series_id = "UNRATE",
  observation_start = as.Date("1990-01-01")
)
#> # A tibble: 342 x 3
#>    date       series_id value
#>    <date>     <chr>     <dbl>
#>  1 1990-01-01 UNRATE      5.4
#>  2 1990-02-01 UNRATE      5.3
#>  3 1990-03-01 UNRATE      5.2
#>  4 1990-04-01 UNRATE      5.4
#>  5 1990-05-01 UNRATE      5.4
#>  6 1990-06-01 UNRATE      5.2
#>  7 1990-07-01 UNRATE      5.5
#>  8 1990-08-01 UNRATE      5.7
#>  9 1990-09-01 UNRATE      5.9
#> 10 1990-10-01 UNRATE      5.9
#> # ... with 332 more rows

Search for FRED series

Use fredr_series_search_text() to search for a series by text in the series description. The data returned is a tibble where each row represents series with description text that matches the text specified by search_text. For example, to search for series with description text that matches “UNRATE”:

fredr_series_search_text(
  search_text = "unemployment",
  limit = 100L
)
#> # A tibble: 100 x 16
#>    id    realtime_start realtime_end title observation_sta… observation_end
#>  * <chr> <chr>          <chr>        <chr> <chr>            <chr>          
#>  1 UNRA… 2018-07-22     2018-07-22   Civi… 1948-01-01       2018-06-01     
#>  2 UNRA… 2018-07-22     2018-07-22   Civi… 1948-01-01       2018-06-01     
#>  3 NROU  2018-07-22     2018-07-22   Natu… 1949-01-01       2027-10-01     
#>  4 UNEM… 2018-07-22     2018-07-22   Unem… 1948-01-01       2018-06-01     
#>  5 M089… 2018-07-22     2018-07-22   Unem… 1929-04-01       1942-06-01     
#>  6 Q089… 2018-07-22     2018-07-22   Unem… 1940-04-01       1946-10-01     
#>  7 M089… 2018-07-22     2018-07-22   Unem… 1940-01-01       1946-12-01     
#>  8 M089… 2018-07-22     2018-07-22   Unem… 1947-01-01       1966-12-01     
#>  9 LNS1… 2018-07-22     2018-07-22   Unem… 1948-01-01       2018-06-01     
#> 10 U6RA… 2018-07-22     2018-07-22   Tota… 1994-01-01       2018-06-01     
#> # ... with 90 more rows, and 10 more variables: frequency <chr>,
#> #   frequency_short <chr>, units <chr>, units_short <chr>,
#> #   seasonal_adjustment <chr>, seasonal_adjustment_short <chr>,
#> #   last_updated <chr>, popularity <int>, group_popularity <int>,
#> #   notes <chr>

Use fredr_series_search_id() to search for a series by character ID of the series. The data returned is a tibble where each row represents series with an ID that matches the text specified by search_text. For example, to search for series with an ID that matches “UNRATE”:

fredr_series_search_id(
  search_text = "UNRATE",
  limit = 100L
)
#> # A tibble: 16 x 16
#>    id    realtime_start realtime_end title observation_sta… observation_end
#>  * <chr> <chr>          <chr>        <chr> <chr>            <chr>          
#>  1 UNRA… 2018-07-22     2018-07-22   Civi… 1948-01-01       2018-06-01     
#>  2 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#>  3 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#>  4 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#>  5 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#>  6 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#>  7 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#>  8 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#>  9 UNRA… 2018-07-22     2018-07-22   Long… 2015-06-17       2018-06-13     
#> 10 UNRA… 2018-07-22     2018-07-22   Civi… 1948-01-01       2018-06-01     
#> 11 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#> 12 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#> 13 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#> 14 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#> 15 UNRA… 2018-07-22     2018-07-22   FOMC… 2018-01-01       2020-01-01     
#> 16 UNRA… 2018-07-22     2018-07-22   Long… 2009-02-18       2018-06-13     
#> # ... with 10 more variables: frequency <chr>, frequency_short <chr>,
#> #   units <chr>, units_short <chr>, seasonal_adjustment <chr>,
#> #   seasonal_adjustment_short <chr>, last_updated <chr>, popularity <int>,
#> #   group_popularity <int>, notes <chr>

Search for FRED series by tags

Use fredr_series_search_tags() to search for series tags. The data returned is a tibble where each row represents a tag matching the text in series_search_text. For example, to return the top 100 tags (by series count) matching the text “unemployment”:

fredr_series_search_tags(
  series_search_text = "unemployment",
  limit = 100L
)
#> # A tibble: 100 x 6
#>    name     group_id notes             created     popularity series_count
#>  * <chr>    <chr>    <chr>             <chr>            <int>        <int>
#>  1 unemplo… gen      ""                2012-02-27…         74        23892
#>  2 nsa      seas     Not Seasonally A… 2012-02-27…        100        23173
#>  3 usa      geo      United States of… 2012-02-27…         99        18376
#>  4 bls      src      Bureau of Labor … 2012-02-27…         85        14408
#>  5 rate     gen      ""                2012-02-27…         87        14216
#>  6 nation   geot     ""                2012-02-27…        100        12647
#>  7 annual   freq     ""                2012-02-27…         91        11127
#>  8 mei      rls      Main Economic In… 2012-08-16…         78        10770
#>  9 oecd     src      Org. for Economi… 2012-02-27…         78        10770
#> 10 county   geot     County or County… 2012-02-27…         76         9504
#> # ... with 90 more rows

Search for FRED series by related tags

Use fredr_search_related_tags() to search for series tags related to a given tag. The data returned is a tibble where each row represents a tag who’s series matches the text in series_search_text and related tags specified by tag_names. For example, to search for tags related to the tag “usa” in which the series text matches “gnp”, use the following:

fredr_series_search_related_tags(
  series_search_text = "gnp",
  tag_names = "usa",
  limit = 30L
)
#> # A tibble: 30 x 6
#>    name    group_id notes             created      popularity series_count
#>  * <chr>   <chr>    <chr>             <chr>             <int>        <int>
#>  1 nation  geot     ""                2012-02-27 …        100           61
#>  2 gnp     gen      Gross National P… 2012-02-27 …         46           46
#>  3 quarte… freq     ""                2012-02-27 …         85           41
#>  4 bea     src      Bureau of Econom… 2012-02-27 …         79           37
#>  5 nipa    rls      National Income … 2012-08-16 …         72           37
#>  6 sa      seas     Seasonally Adjus… 2012-02-27 …         87           36
#>  7 gdp     gen      Gross Domestic P… 2012-02-27 …         83           30
#>  8 nsa     seas     Not Seasonally A… 2012-02-27 …        100           25
#>  9 nber    src      National Bureau … 2012-02-27 …         59           22
#> 10 annual  freq     ""                2012-02-27 …         91           20
#> # ... with 20 more rows

Get basic information for a FRED series

The fredr_series() function returns information for a single series specified by series_id. The data returned is a tibble in which each row represents the series specified. For example, to get information for the UNRATE series:

fredr_series(series_id = "UNRATE")
#> # A tibble: 1 x 15
#>   id    realtime_start realtime_end title observation_sta… observation_end
#> * <chr> <chr>          <chr>        <chr> <chr>            <chr>          
#> 1 UNRA… 2018-07-22     2018-07-22   Civi… 1948-01-01       2018-06-01     
#> # ... with 9 more variables: frequency <chr>, frequency_short <chr>,
#> #   units <chr>, units_short <chr>, seasonal_adjustment <chr>,
#> #   seasonal_adjustment_short <chr>, last_updated <chr>, popularity <int>,
#> #   notes <chr>

Note that there may potentially be more than one row returned if the series has been revised and real time periods are adjusted:

fredr_series(
  series_id = "UNRATE",
  realtime_start = as.Date("1950-01-01")
)
#> # A tibble: 3 x 15
#>   id    realtime_start realtime_end title observation_sta… observation_end
#> * <chr> <chr>          <chr>        <chr> <chr>            <chr>          
#> 1 UNRA… 1960-03-15     1967-02-08   Civi… 1948-01-01       1966-12-01     
#> 2 UNRA… 1967-02-09     2012-02-02   Civi… 1948-01-01       2011-12-01     
#> 3 UNRA… 2012-02-03     9999-12-31   Civi… 1948-01-01       2018-06-01     
#> # ... with 9 more variables: frequency <chr>, frequency_short <chr>,
#> #   units <chr>, units_short <chr>, seasonal_adjustment <chr>,
#> #   seasonal_adjustment_short <chr>, last_updated <chr>, popularity <int>,
#> #   notes <chr>

Get the categories for a FRED series

The fredr_series_categories() function returns a list of categories for the series specified by series_id. The data returned is a tibble in which each row represents a category that the series belongs to. For example, to get the categories for the UNRATE series:

fredr_series_categories(series_id = "UNRATE")
#> # A tibble: 1 x 4
#>      id name         parent_id notes                                      
#> * <int> <chr>            <int> <chr>                                      
#> 1 32447 Unemploymen…        12 The ratio of unemployed to the civilian la…

Get the release for a FRED series

The fredr_series_release() function returns a list of releases that the series specified by series_id belongs to. The data returned is a tibble in which each row represents a release that the series belongs to. For example, to get the release for the UNRATE series:

fredr_series_release(series_id = "UNRATE")
#> # A tibble: 1 x 6
#>      id realtime_start realtime_end name        press_release link        
#> * <int> <chr>          <chr>        <chr>       <lgl>         <chr>       
#> 1    50 2018-07-22     2018-07-22   Employment… TRUE          http://www.…

Get the tags for a FRED series

The fredr_series_tags() function returns a list of tags that are assigned to the series specified by series_id. The data returned is a tibble in which each row represents a tag assigned to the series. For example, to get the tags for the UNRATE series:

fredr_series_tags(
  series_id = "UNRATE",
  order_by = "name"
)
#> # A tibble: 13 x 6
#>    name       group_id notes           created     popularity series_count
#>  * <chr>      <chr>    <chr>           <chr>            <int>        <int>
#>  1 16 years + gen      ""              2012-02-27…         58         5810
#>  2 bls        src      Bureau of Labo… 2012-02-27…         85       113265
#>  3 civilian   gen      ""              2012-02-27…         51         5977
#>  4 headline … gen      ""              2013-11-19…         52           53
#>  5 household… gen      Current Popula… 2012-06-18…         60         2975
#>  6 labor      gen      ""              2012-02-27…         60        24608
#>  7 labor und… gen      ""              2012-05-10…         35          726
#>  8 monthly    freq     ""              2012-02-27…         92       100536
#>  9 nation     geot     ""              2012-02-27…        100       174523
#> 10 rate       gen      ""              2012-02-27…         87        45813
#> 11 sa         seas     Seasonally Adj… 2012-02-27…         87        61222
#> 12 unemploym… gen      ""              2012-02-27…         74        23892
#> 13 usa        geo      United States … 2012-02-27…         99       368894

Get a set of recently updated FRED series

The fredr_series_updates() function returns a list of series recently updated on the FRED server. The data returned is a tibble in which each row represents a series. For example, the default call simply lists 1000 recent updates (the default for the limit parameter), most recent updates appearing first:

fredr_series_updates()
#> # A tibble: 1,000 x 15
#>    id    realtime_start realtime_end title observation_sta… observation_end
#>  * <chr> <chr>          <chr>        <chr> <chr>            <chr>          
#>  1 CBBC… 2018-07-22     2018-07-22   Coin… 2017-12-20       2018-07-22     
#>  2 CBLT… 2018-07-22     2018-07-22   Coin… 2016-08-17       2018-07-22     
#>  3 CBCC… 2018-07-22     2018-07-22   Coin… 2015-01-01       2018-07-21     
#>  4 CBBT… 2018-07-22     2018-07-22   Coin… 2014-12-01       2018-07-22     
#>  5 CBET… 2018-07-22     2018-07-22   Coin… 2016-05-18       2018-07-22     
#>  6 R250… 2018-07-22     2018-07-22   Russ… 1985-12-31       2018-07-20     
#>  7 RTOP… 2018-07-22     2018-07-22   Russ… 1985-12-31       2018-07-20     
#>  8 R250… 2018-07-22     2018-07-22   Russ… 1985-12-31       2018-07-20     
#>  9 RTOP… 2018-07-22     2018-07-22   Russ… 1985-12-31       2018-07-20     
#> 10 RTOP… 2018-07-22     2018-07-22   Russ… 1985-12-31       2018-07-20     
#> # ... with 990 more rows, and 9 more variables: frequency <chr>,
#> #   frequency_short <chr>, units <chr>, units_short <chr>,
#> #   seasonal_adjustment <chr>, seasonal_adjustment_short <chr>,
#> #   last_updated <chr>, popularity <int>, notes <chr>

Use the start_time and end_time parameters to filter the results by time. For example, to get all the macroeconomic times series updated in the last hour:

fredr_series_updates(
  start_time = Sys.time() - 60 * 60,
  end_time = Sys.time(),
  filter_value = "macro"
)
#> # A tibble: 0 x 0

Get the data vintage dates for a FRED series

The fredr_series_vintagedates() function returns a sequence of dates in history when the series specified by series_id was revised or appended to. The data returned is a tibble where each row is a date. For example, to get the vintage dates for the series UNRATE:

fredr_series_vintagedates(series_id = "UNRATE")
#> # A tibble: 701 x 1
#>    vintage_dates
#>    <chr>        
#>  1 1960-03-15   
#>  2 1960-04-12   
#>  3 1960-05-11   
#>  4 1960-06-13   
#>  5 1960-07-12   
#>  6 1960-08-10   
#>  7 1960-09-09   
#>  8 1960-10-10   
#>  9 1960-11-10   
#> 10 1960-12-12   
#> # ... with 691 more rows

FRED Series

Sam Boysel

2018-07-22