sie_banxico

PyPi Version

A python class for the Economic Information System (SIE) API of Banco de México.

Args:
token (str): A query token from Banco de México
id_series (list): A list with the economic series id or with the series id range to query. ** A list must be given even though only one serie is consulted.
language (str): Language of the obtained information. ‘en’ (default) for english or ‘es’ for spanish

Notes:
(1) In order to retrive information from the SIE API, a query token is required. The token can be requested here
(2) Each economic serie is related to an unique ID. The full series catalogue can be consulted here

Pypi Installation

pip install sie_banxico

SIEBanxico Class Instance

Querying Monetary Aggregates M1 (SF311408) and M2 (SF311418) Data

 >>> from api_banxico import SIEBanxico
 >>> api = SIEBanxico(token = token, id_series = ['SF311408' ,'SF311418'], language = 'en')

Class documentation and attributes

>>> api.__doc__
'Returns the full class documentation'
>>> api.token
'1b7da065cf574289a2cb511faeXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX' # This is an example token
>>> api.series
'SF311408,SF311418'

Methods for modify the arguments of the object

set_token: Change the current query token

>>> api.set_token(token = new_token)

set_id_series: Allows to change the series to query

>>> api.append_id_series(id_series = ['SF311412'])
>>> api.series
'SF311408,SF311418,SF311412'

append_id_series: Allows to update the series to query

>>> api.set_id_series(id_series='SF311408-SF311418')
>>> api.series
'SF311408-SF311418'

GET Request Methods

>>> api = SIEBanxico(token = token, id_series = ['SF311408' ,'SF311418']

get_metadata: Allows to consult metadata of the series

    Allows to consult metadata of the series.
    Returns:
        dict: json response format


>>> api.get_metadata()
{'bmx': {'series': [{'idSerie': 'SF311418', 'titulo': 'Monetary Aggregates M2 = M1 + monetary instruments held by residents', 'fechaInicio': '12/01/2000', 'fechaFin': '11/01/2021', 'periodicidad': 'Monthly', 'cifra': 'Stocks', 'unidad': 'Thousands of Pesos', 'versionada': False}, {'idSerie': 'SF311408', 'titulo': 'Monetary Aggregates M1', 'fechaInicio': '12/01/2000', 'fechaFin': '11/01/2021', 'periodicidad': 'Monthly', 'cifra': 'Stocks', 'unidad': 'Thousands of Pesos', 'versionada': False}]}}

get_lastdata: Returns the most recent published data

Returns the most recent published data for the requested series.
Args:
pct_change (str, optional): None (default) for levels, “PorcObsAnt” for change rate compared to the previous observation, “PorcAnual” for anual change rate, “PorcAcumAnual” for annual acummulated change rate.
Returns:
dict: json response format

>>> api.get_lastdata()
{'bmx': {'series': [{'idSerie': 'SF311418', 'titulo': 'Monetary Aggregates M2 = M1 + monetary instruments held by residents', 'datos': [{'fecha': '01/11/2021', 'dato': '11,150,071,721.09'}]}, {'idSerie': 'SF311408', 'titulo': 'Monetary Aggregates M1', 'datos': [{'fecha': '01/11/2021', 'dato': '6,105,266,291.65'}]}]}}

get_timeseries: Allows to consult time series data

    Allows to consult the whole time series data, corresponding to the period defined between the initial date and the final date in the metadata.
    Args:
        pct_change (str, optional): None (default) for levels, "PorcObsAnt" for change rate compared to the previous observation, "PorcAnual" for anual change rate, "PorcAcumAnual" for annual acummulated change rate.
    Returns:
        dict: json response format


>>> api.get_timeseries(pct_change='PorcAnual')
{'bmx': {'series': [{'idSerie': 'SF311418',
    'titulo': 'Monetary Aggregates M2 = M1 + monetary instruments held by residents',
    'datos': [{'fecha': '01/12/2001', 'dato': '12.89'},
     {'fecha': '01/01/2002', 'dato': '13.99'},
     ...
     {'fecha': '01/11/2021', 'dato': '13.38'}],
     'incrementos': 'PorcAnual'}]}}

get_timeseries_range: Returns the data for the period defined

    Returns the data of the requested series, for the defined period.
    Args:
        init_date (str): The date on which the period of obtained data starts. The date must be sent in the format yyyy-mm-dd. If the given date is out of the metadata time range, the oldest value is returned.
        end_date (str): The date on which the period of obtained data concludes. The date must be sent in the format yyyy-mm-dd. If the given date is out of the metadata time range, the most recent value is returned.
        pct_change (str, optional): None (default) for levels, "PorcObsAnt" for change rate compared to the previous observation, "PorcAnual" for anual change rate, "PorcAcumAnual" for annual acummulated change rate.     
    Returns:
        dict: json response format


>>> api.get_timeseries_range(init_date='2000-12-31', end_date='2004-04-01')
{'bmx': {'series': [{'idSerie': 'SF311408',
    'titulo': 'Monetary Aggregates M1',
    'datos': [{'fecha': '01/01/2001', 'dato': '524,836,129.99'},
     {'fecha': '01/02/2001', 'dato': '517,186,605.97'},
     ...
     {'fecha': '01/04/2004', 'dato': '2,306,755,672.89'}]}]}}

Pandas integration for data manipulation (and further analysis)

All the request methods returns a response in json format that can be used with other Python libraries.

The response for the api.get_timeseries_range(init_date=’2000-12-31′, end_date=’2004-04-01′) is a nested dictionary, so we need to follow a path to extract the specific values for the series and then transform the data into a pandas object; like a Serie or a DataFrame. For example:

data = api.get_timeseries_range(init_date='2000-12-31', end_date='2004-04-01')

# Extract the Monetary Aggregate M1 data
data['bmx']['series'][0]['datos']
[{'fecha': '01/01/2001', 'dato': '524,836,129.99'},
 ...
 {'fecha': '01/04/2004', 'dato': '799,774,807.43'}]

# Transform the data into a pandas DataDrame
import pandas as pd
df = pd.DataFrame(timeseries_range['bmx']['series'][0]['datos'])
df.head()
        fecha            dato
0  01/01/2001  524,836,129.99
1  01/02/2001  517,186,605.97
2  01/03/2001  509,701,873.04
3  01/04/2001  511,952,430.01
4  01/05/2001  514,845,459.96

Another useful pandas function to transform json formats into a dataframe is ‘json_normalize’:

df = pd.json_normalize(timeseries_range['bmx']['series'], record_path = 'datos', meta = ['idSerie', 'titulo'])
df['titulo'] = df['titulo'].apply(lambda x: x.replace('Monetary Aggregates M2 = M1 + monetary instruments held by residents', 'Monetary Aggregates M2'))
df.head()
        fecha            dato   idSerie                  titulo
0  01/01/2001  524,836,129.99  SF311408  Monetary Aggregates M1
1  01/02/2001  517,186,605.97  SF311408  Monetary Aggregates M1
2  01/03/2001  509,701,873.04  SF311408  Monetary Aggregates M1
3  01/04/2001  511,952,430.01  SF311408  Monetary Aggregates M1
4  01/05/2001  514,845,459.96  SF311408  Monetary Aggregates M1
df.tail()
         fecha              dato   idSerie                  titulo
75  01/12/2003  2,331,594,974.69  SF311418  Monetary Aggregates M2
76  01/01/2004  2,339,289,328.74  SF311418  Monetary Aggregates M2
77  01/02/2004  2,285,732,239.36  SF311418  Monetary Aggregates M2
78  01/03/2004  2,312,217,167.10  SF311418  Monetary Aggregates M2
79  01/04/2004  2,306,755,672.89  SF311418  Monetary Aggregates M2

Licence

The MIT License (MIT)

By

Dillan Aguirre Sedeño
([email protected])

GitHub

View Github