In Python, ek.get_timeseries() for most Turkish stocks have started to produce wrong historical OPEN/HIGH/LOW/CLOSE values for the period prior to 27/07/2020. Problem started occuring on 19/10/2020.
When I check the CHTV for single Turkish stocks on Reuters Eikon App, same problem is seen there as well. It appears that there is a databse issue, from which both the python api and the CHTV app are pulling wrong prices.
Turkey single stock prices prior to 27/07/2020 seem to have been divided by 100. This may be an unintended consequence of the fact that Borsa Istanbul divided equity index values by 100 on 27/07/2020. While prices for equity indices were divided by 100, prices for single stocks were not.
This is a major issue affecting all scripts written using Eikon.
Any help is appreciated,
I think this is an issue on the backend.
I tested this code for several times, sometime the data return on the same scale 7.xxx but sometime the data are on the different scale 7.xxx vs 0.07xx
I raised a ticket no. 09277217 on your behalf to Refinitiv Content Helpdesk asking them 2 questions:
1. Why the scale of the data jumps a lot
2. Why the result sometimes on 7.xxx scale sometimes on 0.07xx scale
sorry to break this into 2 messages
To give a few examples, in both ek_get_timeseries() and Eikon CHTV app,
a) Closing price of GARAN.IS jumps from 0.08 on July 24th 2020 to 7.02 on July 2th 2020
B) Closing price of ASELS.IS jumps from 0.33 on July 24th 2020 to 35.26 on July 27th 2020
Same problem persists across all single stocks in Turkey. Prices before July 27th 2020 seem to have been divided by 100 across the board and are wrong. There is a huge jump in all single stock prices on July 27th because they revert to normal.
Hope this makes the issue more clear