FRED Wrapper in Python

This repository contains a Python class that allows you to interact with and get data from the FRED website in a simple way without the need to connect to the API or register or get a KEY API.

Installation

Install via pip with the Github repository:

pip install git+https://github.com/Jaldekoa/FRED-Wrapper.git#egg=FRED_Wrapper

You also can install it in the traditional way:

git clone https://github.com/Jaldekoa/FRED-Wrapper.git
cd FRED-Wrapper
python setup.py install

Methods

The Python class provides a way to get data from the FRED web page. Instead of using the official API, this class makes direct requests to the web page and gets the data in CSV format.

get_fred_data()

This method allows obtaining single series as well as multiple series of FRED data. The data can be customized using various parameters, such as series ID, start date, end date, data transformation, frequency, aggregation method and applied formula. The function returns the data in the form of a pandas DataFrame, which facilitates its manipulation and analysis.

  FRED_Wrapper.get_fred_data(**kwargs)

Keyword Args:

Parameter	Type	Description
id	str or list[str]	ID or key of the requested data series.
start_date	str or list[str]	Start date of the data series. Format: YYYY-MM-DD
end_date	str or list[str]	Start date of the data series. Format: YYYY-MM-DD
transform	str or list[str]	Transformation of the data series. Valid values: None: "lin", Change: "chg", Change from Year Ago: "ch1", Percent Change from Year Ago: "pc1", Compounded Annual Rate of Change: "pca", Continuously Compounded Rate of Change: "cch", Continuously Compounded Annual Rate of Change: "cca".
freq	str or list[str]	Frequency of the data series.
agg	str or list[str]	Aggregation method of the data series. Valid values: Average: "avg", Sum: "sum", End of Period: "eop".
formula	str or list[str]	Formula applied to the data series.

Returns:

• pd.DataFrame: A DataFrame containing series data.

Usage/Examples

Single series

You can obtain the data for a single series as follows:

from FRED_Wrapper import FRED_Wrapper

df = FRED_Wrapper.get_fred_data(id="WM1NS",
				start_date="2020-01-01",
				end_date="2023-08-31",
				transform="chg",
				freq="Monthly",
				agg="eop",
				formula="a/1000")

Result

print(df.tail())

	DATE	WM1NS_CHG
39	2023-04-01 00:00:00	-0.3139
40	2023-05-01 00:00:00	-0.08
41	2023-06-01 00:00:00	-0.1384
42	2023-07-01 00:00:00	-0.0522
43	2023-08-01 00:00:00	-0.1316

Multiple series

You can obtain the data from a multiple series in either of the following two ways:

from FRED_Wrapper import FRED_Wrapper

# First way
df = FRED_Wrapper.get_fred_data(id=["WM1NS", "WM2NS"],
				start_date=["2020-01-01", "2020-01-01"],
				end_date=["2023-08-31", "2023-08-31"],
				transform=["chg", "chg"],
				freq=["Monthly", "Monthly"],
				agg=["eop", "eop"],
				formula=["a/1000", "a/1000"])
)

from FRED_Wrapper import FRED_Wrapper

# Second way
PARAMS = {
    "id": ["WM1NS", "WM2NS"],  
    "start_date": ["2020-05-01", "2020-05-01"],  
    "end_date": ["2023-08-31", "2023-08-31"],
    "transform": ["chg", "chg"],  
    "freq": ["Monthly", "Monthly"],  
    "agg": ["eop", "eop"],  
    "formula": ["a/1000", "a/1000"],  
}  
  
df = FRED_Wrapper.get_fred_data(**PARAMS)

You are not restricted to getting only two series at a time. You can get as many series as you want in a single method call. The only condition is that it will return all series in a single pandas DataFrame.

Result

print(df.tail())

	DATE	WM1NS_CHG	WM2NS_CHG
39	2023-04-01 00:00:00	-0.3139	-0.1962
40	2023-05-01 00:00:00	-0.08	0.081
41	2023-06-01 00:00:00	-0.1384	-0.0595
42	2023-07-01 00:00:00	-0.0522	0.0354
43	2023-08-01 00:00:00	-0.1316	-0.0562