Get Statistical Time Series for a Place

datacommons.get_stat_series(place, stat_var, measurement_method=None,observation_period=None, unit=None, scaling_factor=None)

Returns a dict mapping date to value for a place based on the stat_var, with optional filter parameters.

See the full list of StatisticalVariables.


  • place (str): The dcid of the Place to query for.

  • stat_var (str): The dcid of the StatisticalVariable.

  • measurement_method (str): (Optional) The dcid of the preferred measurementMethod for the stat_var.

  • observation_period (str): (Optional) The preferred observationPeriod for the stat_var. This is an ISO 8601 duration such as P1M (one month).

  • unit (str): (Optional) The dcid of the preferred unit for the stat_var.

  • scaling_factor (int): (Optional) The preferred scalingFactor for the stat_var.


A dict mapping date(str) to the statistical value (float).


  • ValueError - If no statistical value found for the place with the given parameters.

Be sure to initialize the library. Check the Python library setup guide for more details.

You can find a list of StatisticalVariables with human-readable names here.


We would like to get the male population in Arkansas

>>> import datacommons as dc
>>> dc.get_stat_series("geoId/05", "Count_Person_Male")
  "2013": 1439862,
  "2014": 1447235,
  "2015": 1451913,
  "2016": 1456694,
  "2017": 1461651,
  "2018": 1468412,
  "2011": 1421287,
  "2012": 1431252

In the next example, the parameter observation_period='P3Y' overly constrains the request so the API throws ValueError:

>>> dc.get_stat_series('geoId/06085', 'Count_Person', observation_period='P3Y')
Traceback (most recent call last):
    raise ValueError('No data in response.')