Notebook
In [1]:
from quantopian.interactive.data.quandl import fred_gdp as dataset
from quantopian.interactive.data.quandl import fred_gdp_deltas as datasetd
In [2]:
#initial value GDP
dataset[(dataset.asof_date == '2015-10-01')]
Out[2]:
value asof_date timestamp
0 18128.2 2015-10-01 2016-01-30 04:02:06.493524
In [3]:
#updated records GDP (18287.226 from datareader) Timestamp is the datetime we noticed the change
datasetd[(datasetd.asof_date == '2015-10-01')]
Out[3]:
value asof_date timestamp
0 18148.400 2015-10-01 2016-02-27 06:03:52.892209
1 18164.800 2015-10-01 2016-03-26 07:03:59.669764
2 18222.800 2015-10-01 2016-07-29 16:05:04.999148
3 18287.200 2015-10-01 2017-07-29 03:06:37.777611
4 18287.226 2015-10-01 2017-10-30 19:07:11.118510
In [4]:
from quantopian.interactive.data.quandl import fred_retailsmnsa_deltas as datasetd
from quantopian.interactive.data.quandl import fred_retailsmnsa as dataset
In [5]:
#initial value
dataset[(dataset.asof_date == '2015-09-01')]
Out[5]:
value asof_date timestamp
0 379840.0 2015-09-01 2015-09-02
In [6]:
#updated records REATAILSMNSA (382242 from datareader)
datasetd[(datasetd.asof_date == '2015-09-01')]
Out[6]:
value asof_date timestamp
0 379473.0 2015-09-01 2015-12-14 20:33:56.456006
1 380320.0 2015-09-01 2016-05-13 16:01:38.674138
2 382242.0 2015-09-01 2017-05-14 03:09:43.234678

results from DataReader

import pandas_datareader.data as web  
import datetime
my_data = web.DataReader(['GDP','RETAILSMNSA'], 'fred', datetime.datetime(2015, 9, 1) , datetime.datetime(2015, 10, 2))
my_data
DATE          GDP         RETAILSMNSA   
2015-09-01  NaN         382242
2015-10-01  18287.226     392786

These match the last timestamp in the deltas tables above