Skip to content
/ beaapi Public

A Python library library to make it easier to retrieve and work with BEA data.

License

CC0-1.0, Unknown licenses found

Licenses found

CC0-1.0
LICENSE
Unknown
LICENSE.note
Notifications You must be signed in to change notification settings

us-bea/beaapi

beaapi

beaapi: A Python library library to make it easier to retrieve and work with BEA data. For the parallel R-package, see bea.R.

To install the package, download the .whl file for the latest release, and install using pip.

For help, see our documentation.

For development and building information, see CONTRIBUTING.md.

To Get Started

Once you have installed the package, you can load it in your Python script as follows:

import beaapi

To use the package, you must first register for an API key from BEA by providing your name and email address. The key will be emailed to you.

Once you have received your BEA API key, you can save it to a variable to make it easier to use later:

beakey = 'YOUR 36-DIGIT API KEY'

For scripts, an even better method is to create an unversioned text file called .env with the contents beakey=Your36DigitAPiKey and then use the package python-dotenv to load it automatically.

import os
from dotenv import load_dotenv
load_dotenv()
beakey = os.environ.get("beakey")

Get metadata

The BEA data is distributed across various datasets that each contain many tables. To get a list of the datasets available on the API, use get_data_set_list():

list_of_sets = beaapi.get_data_set_list(beakey)
display(list_of_sets)  # Note the last dataset is only for speeding up metadata queries
DatasetName DatasetDescription
0 NIPA Standard NIPA tables
1 NIUnderlyingDetail Standard NI underlying detail tables
2 MNE Multinational Enterprises
3 FixedAssets Standard Fixed Assets tables
4 ITA International Transactions Accounts
5 IIP International Investment Position
6 InputOutput Input-Output Data
7 IntlServTrade International Services Trade
8 GDPbyIndustry GDP by Industry
9 Regional Regional data sets
10 UnderlyingGDPbyIndustry Underlying GDP by Industry
11 APIDatasetMetaData Metadata about other API datasets

Queries to the different datasets take different parameters. To get a list of the paramaters for a given dataset, use get_parameter_list(); for example, to get a list of the parameters for the NIPA dataset, use:

list_of_params = beaapi.get_parameter_list(beakey, 'NIPA')
display(list_of_params)
ParameterName ParameterDataType ParameterDescription ParameterIsRequiredFlag ParameterDefaultValue MultipleAcceptedFlag AllValue
0 Frequency string A - Annual, Q-Quarterly, M-Monthly 1 1
1 ShowMillions string A flag indicating that million-dollar data sho... 0 N 0
2 TableID integer The standard NIPA table identifier 0 <NA> 0
3 TableName string The new NIPA table identifier 0 <NA> 0
4 Year integer List of year(s) of data to retrieve (X for All) 1 1 X

To get a list of the values for a given parameter, use get_parameter_values(); for example, to get a list of the parameter values for the Frequency parameter of the NIPA dataset, use:

list_of_param_vals = beaapi.get_parameter_values(beakey, 'NIPA', 'Frequency')
display(list_of_param_vals)
FrequencyID Description
0 A Annual
1 Q Quarterly
2 M Monthly

The API documentation is a good place to also understand allowed values.

A few datasets (ITA, IIP, InputOutput, IntlServTrade, GDPbyIndustry, Regional, and UnderlyingGDPbyIndustry) allow you to filter the values based on what will be passed in for other parameters.

tbl = beaapi.get_parameter_values_filtered(beakey, 'ITA', targetparameter='Indicator', AreaOrCountry="China",Frequency="A", Year="2011")
display(tbl.head())
Key Desc
0 BalCapAcct Balance on capital account
1 BalCurrAcct Balance on current account
2 BalGds Balance on goods
3 BalGdsServ Balance on goods and services
4 BalPrimInc Balance on primary income

A few of the datasets publish metadata tables that can be queried for particular strings using search_metadata(). This method allows you to search for BEA data by keyword. For example, to find all datasets in which the term "personal consumption" appears, use the following:

search_data = beaapi.search_metadata('Gross domestic', beakey)
search_data.head(2)
SeriesCode RowNumber LineDescription LineNumber ParentLineNumber Tier Path TableId Datasetname TableName ReleaseDate NextReleaseDate
0 A191RL 10 Gross domestic product 1 0 1 T10101 NIPA Table 1.1.1. Percent Change From Preceding Per... Feb 28 2019 8:30AM Mar 28 2019 8:30AM
26 A191RP 280 Gross domestic product, current dollars 27 0 27 T10101 NIPA Table 1.1.1. Percent Change From Preceding Per... Feb 28 2019 8:30AM Mar 28 2019 8:30AM

Please note that that search_metadata currently searches only national data.

The contents of this function are automatically updated using a new metadata component of BEA's API; as such, we recommend that you use it with your API key, and the first use of this function requires that you use your key or it will be unable to extract the metadata.

If you do not wish to automatically update the metadata (e.g., you have conducted a study using the search function), simply searching for the term without also passing your key to the function will do a search only using your locally stored version.

Get data

Once you have identified the TableId number and other information, you can use get_data() to access the data. The following code, for example, returns the NIPA table with 2015 data for Table T20305.

bea_tbl = beaapi.get_data(beakey, datasetname='NIPA', TableName='T20305', Frequency='Q', Year='2015')
display(bea_tbl.head(2))
TableName SeriesCode LineNumber LineDescription TimePeriod METRIC_NAME CL_UNIT UNIT_MULT DataValue NoteRef
0 T20305 DPCERC 1 Personal consumption expenditures (PCE) 2015Q1 Current Dollars Level 6 12083904 T20305
1 T20305 DPCERC 1 Personal consumption expenditures (PCE) 2015Q2 Current Dollars Level 6 12224707 T20305

We store in the meta-data the index columns so that you can create a unique index on the data-frame.

display(bea_tbl.set_index(bea_tbl.attrs['index_cols']).head(2))
TableName SeriesCode LineDescription METRIC_NAME CL_UNIT UNIT_MULT DataValue NoteRef
LineNumber TimePeriod
1 2015Q1 T20305 DPCERC Personal consumption expenditures (PCE) Current Dollars Level 6 12083904 T20305
2015Q2 T20305 DPCERC Personal consumption expenditures (PCE) Current Dollars Level 6 12224707 T20305

Extra meta-data from the API is returned in a dictionary in the attributes called detail and can vary based on the dataset.

print('Extra detail keys:' + str(bea_tbl.attrs['detail'].keys()))
print("Let's look at some interesting ones.")
print('Statistic: ' + bea_tbl.attrs['detail']['Statistic'])
print("Notes corresponding to NoteRef:")
display(bea_tbl.attrs['detail']['Notes'].head())
Extra detail keys:dict_keys(['Statistic', 'UTCProductionTime', 'Dimensions', 'Notes'])
Let's look at some interesting ones.
Statistic: NIPA Table
Notes corresponding to NoteRef:
NoteRef NoteText
0 T20305 Table 2.3.5. Personal Consumption Expenditures...
1 T20305.1 1. Net expenses of NPISHs, defined as their gr...
2 T20305.2 2. Gross output is net of unrelated sales, sec...
3 T20305.3 3. Excludes unrelated sales, secondary sales, ...
4 T20305.4 4. Food consists of food and beverages purchas...

To retrieve a limited selection of multiple years, list all the years you want to retrieve. For example, to retrieve data for 2011-2015, use Year="2011,2012,2013,2014,2015"

The API documentation includes information about the specific parameters required by get_data().

If you would like to format the data in a wide format so that different variables (in this case LineNumbers) are different rows, you can use to_wide_vars_to_cols().

display(beaapi.to_wide_vars_in_cols(bea_tbl).head(3))
LineNumber 1 2 3 4 5 6 7 8 9 10 ... 19 20 21 22 23 24 25 26 27 28
TimePeriod
2015Q1 12083904 3867908 1283823 465426 285887 332965 199546 2584085 917371 366171 ... 814298 960760 992461 362114 1367850 1005736 10619844 546689 10570891 9107467
2015Q2 12224707 3927357 1309627 481820 292631 333600 201577 2617730 916930 368315 ... 828712 977198 1003757 363147 1374002 1010855 10759482 548295 10693235 9228648
2015Q3 12347752 3960436 1318493 481620 297306 337083 202483 2641944 924494 370385 ... 836130 978430 1012278 373825 1385653 1011828 10877903 545354 10802962 9333737

3 rows × 28 columns

Advanced topics

Throttling: The BEA api limits requests to a maximum of 100/minute and 100MB/minute (as well as 30 errors/minute). If the user exceeds this, they will be denied access for 1 hour. This package will automatically self-throttle, so in general, the user does not have to worry about that.

See the docs for additional information on:

  • How to construct API queries to pull data from existing iTables.

Quick Links

Disclaimer

The United States Department of Commerce (DOC) GitHub project code is provided on an ‘as is’ basis and the user assumes responsibility for its use. DOC has relinquished control of the information and no longer has responsibility to protect the integrity, confidentiality, or availability of the information. Any claims against the Department of Commerce stemming from the use of its GitHub project will be governed by all applicable Federal law. Any reference to specific commercial products, processes, or services by service mark, trademark, manufacturer, or otherwise, does not constitute or imply their endorsement, recommendation or favoring by the Department of Commerce. The Department of Commerce seal and logo, or the seal and logo of a DOC bureau, shall not be used in any manner to imply endorsement of any commercial product or activity by DOC or the United States Government.

Use of this library may result in data being stored on users' local machines. Specifically, local copies of BEA API metadata may be stored and automatically updated in the beaapi_data directory (unless otherwise changed) in order to improve performance of search_metadata().

About

A Python library library to make it easier to retrieve and work with BEA data.

Resources

License

CC0-1.0, Unknown licenses found

Licenses found

CC0-1.0
LICENSE
Unknown
LICENSE.note

Code of conduct

Security policy

Stars

Watchers

Forks