Issues pulling historical MSCI data in Eikon API

sami01

I have an issue pulling in data from MSCI World and MSCI North America from 1969 and forward.

I can easily get the data from 2016 till today, but I need the entire history

If I use the ek.get_timeseries() function it seems that I am limited in the amount of data I can retrieve.

If I on the other hand use the ek.get_data() function, I get no data from 1969 even though the Datastream Navigator says that there should be data dating back to 31/12/1969.

Are you able to assist on this?

raksina.samasiri

hi @sami01 ,

According to Eikon Data API Usage and Limits Guideline, call-based limit of get_timeseries is 3,000 data points (rows) for interday intervals. This limit applies to the whole request, whatever the number of the requested instruments.

So for 2 instruments, the limit of get_timeseries is 1500 (from 3000/2) rows per call. You may call the get_timeseries function as a batch of the 1500 dates per call. The code below can be used to generate the date-ranges between start and end dates

days = 1500
start_date = datetime.datetime.strptime('1969-12-31', '%Y-%m-%d')
end_date = datetime.datetime.strptime('2000-01-01', '%Y-%m-%d')

date_list = []
start_range = start_date
while start_range < end_date:
    end_range = start_range + datetime.timedelta(days=days)
    if end_range > end_date:
        end_range = end_date
    start_end = [start_range.strftime('%Y-%m-%d'), (end_range).strftime('%Y-%m-%d')]
    date_list.append(start_end)
    start_range += datetime.timedelta(days=days + 1)
date_list

Then you can call get_timeseries for each date range, for example,

import numpy as np

columns = ['Date']
for ric in rics:
    columns.append(ric)
dfs = pd.DataFrame(columns = columns).set_index('Date')
dfs.columns.name = 'CLOSE'
for dates in date_list:
    df = ek.get_timeseries(rics, fields='CLOSE',start_date=dates[0], end_date=dates[1])
    # if .dMINA00000PUS' have no data on the requested date range, restructure the dataframe so it can be appended properly
    if len(df.columns) == 1:
        df = df.rename(columns={'CLOSE':df.columns.name})
        df.columns.name = 'CLOSE'
        df['.dMINA00000PUS'] = np.nan
    dfs = dfs.append(df)

display(dfs)

Hope this could help