Pull data from Federal Reserve Economic Data (FRED) directly into Julia



A third-party Julia library to pull data from Federal Reserve Economic Data (FRED).

Does this thing work?
Package Evaluator
Build Status


FredData is not affiliated in any way with Federal Reserve Bank of St. Louis and is not officially maintained or otherwise supported by Federal Reserve Bank of St. Louis.

FredData is free software and is issued under the MIT license.


FredData uses FRED's Developer API. As such, you must register an API key here in order to pull from the FRED servers.

Download the package with

julia> Pkg.add("FredData")

Make the FRED API key that you just registered accessible to FredData in one of several ways. Ideally, we store your key such that it persists across sessions. In subsequent sections, we'll assume that you have stored your key in one of these ways such that it can be detected automatically. This will allow the use of the zero-argument constructor.

  1. Populate a configuration file ~/.freddatarc. julia> open(joinpath(homedir(), ".freddatarc"), "w") do f write(f, "0123456789abcdef0123456789abcdef") end
  2. Populate the environment variable FRED_API_KEY such that it remains across sessions.

    # on macOS/Linux
    shell> echo "export FRED_API_KEY=0123456789abcdef0123456789abcdef" >> ~/.bashrc
    # on Windows 7+
    shell> setx FRED_API_KEY 0123456789abcdef0123456789abcdef

Another option, though less user-friendly, is to provide your API key to the constructor every time you wish to use the package.

  1. Provide the Fred constructor with your API key directly. julia> f = Fred("0123456789abcdef0123456789abcdef") FRED API Connection url: https://api.stlouisfed.org/fred/ key: 0123456789abcdef0123456789abcdef


Basic Usage

Query observations and metadata.

julia> using FredData

julia> f = Fred()
API key loaded.
FRED API Connection
        url: http://api.stlouisfed.org/fred/
        key: 0123456789abcdef0123456789abcdef

julia> a = get_data(f, "GDPC1")
        id: GDPC1
        title: Real Gross Domestic Product
        units: Billions of Chained 2009 Dollars
        seas_adj (native): Seasonally Adjusted Annual Rate
        freq (native): Quarterly
        realtime_start: 2015-12-11
        realtime_end: 2015-12-11
        last_updated: 2015-11-24T08:01:09
        notes: BEA Account Code: A191RX1 Real gross domestic product is the inflation adjusted value of the goods and services produced by labor and property located in the United States. For more information see the Guide to the National Income and Product Accounts of the United States (NIPA) - (http://www.bea.gov/national/pdf/nipaguid.pdf)
        trans_short: lin
        df: 275x4 DataFrame with columns [:realtime_start,:realtime_end,:date,:value]

Advanced Usage

Add optional arguments. All optional arguments specified by the FRED API are supported.

using FredData
f = Fred()
data = get_data(f, "GDPC1"; vintage_dates="2008-09-15")
data = get_data(f, "GDPC1"; frequency="a", units="chg")

For a full list of optional arguments, see ?get_data or here

The Fred type

The Fred type represents a connection to the FRED API.

Get and set fields.

  • get_api_key(f::Fred): Get the base URL used to connect to the server
  • get_api_url(f::Fred): Get the base URL used to connect to the server
  • set_api_url!(f::Fred, url::AbstractString): Set the base URL used to connect to the server

The FredSeries type

The FredSeries type contains the data in a query response.

Get fields.

  • id(s::FredSeries): Series ID
  • title(s::FredSeries): Series title
  • units_short(s::FredSeries): Units (abbr.)
  • units(s::FredSeries): Units
  • seas_adj_short(s::FredSeries): Seasonal adjustment (abbr.)
  • seas_adj(s::FredSeries): Seasonal adjustment
  • freq_short(s::FredSeries): Native frequency (abbr.)
  • freq(s::FredSeries): Native frequency
  • realtime_start(s::FredSeries): Date realtime period starts
  • realtime_end(s::FredSeries): Date realtime period ends
  • last_updated(s::FredSeries): Date series last updated
  • notes(s::FredSeries): Series notes
  • trans_short(s::FredSeries): Transformation of queried data (abbr.)
  • df(s::FredSeries): The actual data; DataFrame with columns :realtime_start, :realtime_end, :date, :value


Todo list:
☐ improve test coverage
☐ support creation of pseudo-vintages
☐ support methods to query other parts of the API, such as releases, tags, and search
☐ improve some API elements to be more idiomatic

First Commit


Last Touched

10 days ago


49 commits

Used By: