Mutual fund past portfolio holdings via Eikon API
Hello,
as a follow up to my recent question here, which was wonderfully answered, I was wondering if it is possible to get the historical/past portfolio holdings of a mutual fund via the Eikon API?
In Eikon, this data is available when viewing a specific fund, then going holdings => derived holdings. A screenshot of that screen is attached, which shows that one can view the past portfolio for any of the past months.
But how would I access this via Python? In the Excel formula builder, I was not able to see a way to enter a date for the fields that return the portfolio holdings, e.g. 'TR.FundHoldingName', 'TR.FundPercentageOfFundAssets', 'TR.FundNumberOfShares'.
Can the following code, which returns the top 100 holdings of a specific fund for the last available date, be adapted to get historical portfolios?
data , err = ek.get_data('LP40065886',
['TR.FundHoldingRIC',
'TR.FundHoldingName',
'TR.FundPercentageOfFundAssets',
'TR.FundNumberOfShares],
{'Endnum':'100'})
I'm afraid I don't see a way of retrieving this data using Eikon Data APIs. I have opened a case on your behalf with Refinitiv Helpdesk to check if there's a way to retrieve this data into Excel using =TR worksheet function. If the Helpdesk produce a method of retrieving this data using =TR function, then we should be able to replicate the same call using Eikon Data APIs. Refinitiv Helpdesk will be in touch with you, but just in case and for your reference the case number is 07609916.
The Helpdesk confirmed that currently historical fund holdings are not available through Excel or through Eikon Data APIs.
Hi Alex,
do you know if anything has changed in the meantime? Is it now possible to get also historical fund holdings data via the API? The data is clearly available in Eikon, it is just a matter of making it accessible via the API.
I think it would be great to have this possibility, because right now I have to pull the current holdings data every month for a year to create a year-long panel data set, whereas it should be possible to build such a data set in a day just by getting current and historical data. Thanks!